Blame view
fs/exec.c
47 KB
457c89965
|
1 |
// SPDX-License-Identifier: GPL-2.0-only |
1da177e4c
|
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 |
/* * 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 |
613cc2b6f
|
23 |
* formats. |
1da177e4c
|
24 |
*/ |
1da177e4c
|
25 26 |
#include <linux/slab.h> #include <linux/file.h> |
9f3acc314
|
27 |
#include <linux/fdtable.h> |
ba92a43db
|
28 |
#include <linux/mm.h> |
615d6e875
|
29 |
#include <linux/vmacache.h> |
1da177e4c
|
30 31 |
#include <linux/stat.h> #include <linux/fcntl.h> |
ba92a43db
|
32 |
#include <linux/swap.h> |
74aadce98
|
33 |
#include <linux/string.h> |
1da177e4c
|
34 |
#include <linux/init.h> |
6e84f3152
|
35 |
#include <linux/sched/mm.h> |
f7ccbae45
|
36 |
#include <linux/sched/coredump.h> |
3f07c0144
|
37 |
#include <linux/sched/signal.h> |
6a3827d75
|
38 |
#include <linux/sched/numa_balancing.h> |
299300258
|
39 |
#include <linux/sched/task.h> |
ca5b172bd
|
40 |
#include <linux/pagemap.h> |
cdd6c482c
|
41 |
#include <linux/perf_event.h> |
1da177e4c
|
42 43 44 45 46 |
#include <linux/highmem.h> #include <linux/spinlock.h> #include <linux/key.h> #include <linux/personality.h> #include <linux/binfmts.h> |
1da177e4c
|
47 |
#include <linux/utsname.h> |
84d737866
|
48 |
#include <linux/pid_namespace.h> |
1da177e4c
|
49 50 |
#include <linux/module.h> #include <linux/namei.h> |
1da177e4c
|
51 52 53 |
#include <linux/mount.h> #include <linux/security.h> #include <linux/syscalls.h> |
8f0ab5147
|
54 |
#include <linux/tsacct_kern.h> |
9f46080c4
|
55 |
#include <linux/cn_proc.h> |
473ae30bc
|
56 |
#include <linux/audit.h> |
6341c393f
|
57 |
#include <linux/tracehook.h> |
5f4123be3
|
58 |
#include <linux/kmod.h> |
6110e3abb
|
59 |
#include <linux/fsnotify.h> |
5ad4e53bd
|
60 |
#include <linux/fs_struct.h> |
3d5992d2a
|
61 |
#include <linux/oom.h> |
0e028465d
|
62 |
#include <linux/compat.h> |
b44a7dfc6
|
63 |
#include <linux/vmalloc.h> |
1da177e4c
|
64 |
|
7c0f6ba68
|
65 |
#include <linux/uaccess.h> |
1da177e4c
|
66 |
#include <asm/mmu_context.h> |
b6a2fea39
|
67 |
#include <asm/tlb.h> |
43d2b1132
|
68 69 |
#include <trace/events/task.h> |
a6f76f23d
|
70 |
#include "internal.h" |
1da177e4c
|
71 |
|
4ff16c25e
|
72 |
#include <trace/events/sched.h> |
d6e711448
|
73 |
int suid_dumpable = 0; |
e4dc1b14d
|
74 |
static LIST_HEAD(formats); |
1da177e4c
|
75 |
static DEFINE_RWLOCK(binfmt_lock); |
8fc3dc5a3
|
76 |
void __register_binfmt(struct linux_binfmt * fmt, int insert) |
1da177e4c
|
77 |
{ |
8fc3dc5a3
|
78 |
BUG_ON(!fmt); |
92eaa565a
|
79 80 |
if (WARN_ON(!fmt->load_binary)) return; |
1da177e4c
|
81 |
write_lock(&binfmt_lock); |
74641f584
|
82 83 |
insert ? list_add(&fmt->lh, &formats) : list_add_tail(&fmt->lh, &formats); |
1da177e4c
|
84 |
write_unlock(&binfmt_lock); |
1da177e4c
|
85 |
} |
74641f584
|
86 |
EXPORT_SYMBOL(__register_binfmt); |
1da177e4c
|
87 |
|
f6b450d48
|
88 |
void unregister_binfmt(struct linux_binfmt * fmt) |
1da177e4c
|
89 |
{ |
1da177e4c
|
90 |
write_lock(&binfmt_lock); |
e4dc1b14d
|
91 |
list_del(&fmt->lh); |
1da177e4c
|
92 |
write_unlock(&binfmt_lock); |
1da177e4c
|
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); } |
90f8572b0
|
101 102 103 104 105 |
bool path_noexec(const struct path *path) { return (path->mnt->mnt_flags & MNT_NOEXEC) || (path->mnt->mnt_sb->s_iflags & SB_I_NOEXEC); } |
69369a700
|
106 |
#ifdef CONFIG_USELIB |
1da177e4c
|
107 108 109 110 111 112 |
/* * 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
|
113 |
SYSCALL_DEFINE1(uselib, const char __user *, library) |
1da177e4c
|
114 |
{ |
72c2d5319
|
115 |
struct linux_binfmt *fmt; |
964bd1836
|
116 |
struct file *file; |
91a27b2a7
|
117 |
struct filename *tmp = getname(library); |
964bd1836
|
118 |
int error = PTR_ERR(tmp); |
47c805dc2
|
119 120 |
static const struct open_flags uselib_flags = { .open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, |
62fb4a155
|
121 |
.acc_mode = MAY_READ | MAY_EXEC, |
f9652e10c
|
122 123 |
.intent = LOOKUP_OPEN, .lookup_flags = LOOKUP_FOLLOW, |
47c805dc2
|
124 |
}; |
964bd1836
|
125 |
|
6e8341a11
|
126 127 |
if (IS_ERR(tmp)) goto out; |
f9652e10c
|
128 |
file = do_filp_open(AT_FDCWD, tmp, &uselib_flags); |
6e8341a11
|
129 130 131 |
putname(tmp); error = PTR_ERR(file); if (IS_ERR(file)) |
1da177e4c
|
132 133 134 |
goto out; error = -EINVAL; |
496ad9aa8
|
135 |
if (!S_ISREG(file_inode(file)->i_mode)) |
1da177e4c
|
136 |
goto exit; |
30524472c
|
137 |
error = -EACCES; |
90f8572b0
|
138 |
if (path_noexec(&file->f_path)) |
6146f0d5e
|
139 |
goto exit; |
1da177e4c
|
140 |
|
2a12a9d78
|
141 |
fsnotify_open(file); |
6110e3abb
|
142 |
|
1da177e4c
|
143 |
error = -ENOEXEC; |
1da177e4c
|
144 |
|
72c2d5319
|
145 146 147 148 149 150 |
read_lock(&binfmt_lock); list_for_each_entry(fmt, &formats, lh) { if (!fmt->load_shlib) continue; if (!try_module_get(fmt->module)) continue; |
1da177e4c
|
151 |
read_unlock(&binfmt_lock); |
72c2d5319
|
152 153 154 155 156 |
error = fmt->load_shlib(file); read_lock(&binfmt_lock); put_binfmt(fmt); if (error != -ENOEXEC) break; |
1da177e4c
|
157 |
} |
72c2d5319
|
158 |
read_unlock(&binfmt_lock); |
6e8341a11
|
159 |
exit: |
1da177e4c
|
160 161 162 |
fput(file); out: return error; |
1da177e4c
|
163 |
} |
69369a700
|
164 |
#endif /* #ifdef CONFIG_USELIB */ |
1da177e4c
|
165 |
|
b6a2fea39
|
166 |
#ifdef CONFIG_MMU |
ae6b585ee
|
167 168 169 170 171 172 |
/* * The nascent bprm->mm is not visible until exec_mmap() but it can * use a lot of memory, account these pages in current->mm temporary * for oom_badness()->get_mm_rss(). Once exec succeeds or fails, we * change the counter back via acct_arg_size(0). */ |
0e028465d
|
173 |
static void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) |
3c77f8457
|
174 175 176 177 178 179 180 181 |
{ struct mm_struct *mm = current->mm; long diff = (long)(pages - bprm->vma_pages); if (!mm || !diff) return; bprm->vma_pages = pages; |
3c77f8457
|
182 |
add_mm_counter(mm, MM_ANONPAGES, diff); |
3c77f8457
|
183 |
} |
0e028465d
|
184 |
static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, |
b6a2fea39
|
185 186 187 188 |
int write) { struct page *page; int ret; |
9beae1ea8
|
189 |
unsigned int gup_flags = FOLL_FORCE; |
b6a2fea39
|
190 191 192 |
#ifdef CONFIG_STACK_GROWSUP if (write) { |
d05f3169c
|
193 |
ret = expand_downwards(bprm->vma, pos); |
b6a2fea39
|
194 195 196 197 |
if (ret < 0) return NULL; } #endif |
9beae1ea8
|
198 199 200 |
if (write) gup_flags |= FOLL_WRITE; |
1e9877902
|
201 202 203 204 |
/* * We are doing an exec(). 'current' is the process * doing the exec and bprm->mm is the new process's mm. */ |
9beae1ea8
|
205 |
ret = get_user_pages_remote(current, bprm->mm, pos, 1, gup_flags, |
5b56d49fc
|
206 |
&page, NULL, NULL); |
b6a2fea39
|
207 208 |
if (ret <= 0) return NULL; |
655c16a8c
|
209 210 |
if (write) acct_arg_size(bprm, vma_pages(bprm->vma)); |
b6a2fea39
|
211 212 213 214 215 216 217 218 |
return page; } static void put_arg_page(struct page *page) { put_page(page); } |
b6a2fea39
|
219 220 221 222 223 224 225 226 227 228 229 230 |
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
|
231 |
int err; |
b6a2fea39
|
232 233 |
struct vm_area_struct *vma = NULL; struct mm_struct *mm = bprm->mm; |
490fc0538
|
234 |
bprm->vma = vma = vm_area_alloc(mm); |
b6a2fea39
|
235 |
if (!vma) |
eaccbfa56
|
236 |
return -ENOMEM; |
bfd40eaff
|
237 |
vma_set_anonymous(vma); |
b6a2fea39
|
238 |
|
f268dfe90
|
239 240 241 242 |
if (down_write_killable(&mm->mmap_sem)) { err = -EINTR; goto err_free; } |
b6a2fea39
|
243 244 245 246 247 248 249 |
/* * 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. */ |
aacb3d17a
|
250 |
BUILD_BUG_ON(VM_STACK_FLAGS & VM_STACK_INCOMPLETE_SETUP); |
b6a2fea39
|
251 252 |
vma->vm_end = STACK_TOP_MAX; vma->vm_start = vma->vm_end - PAGE_SIZE; |
d9104d1ca
|
253 |
vma->vm_flags = VM_SOFTDIRTY | VM_STACK_FLAGS | VM_STACK_INCOMPLETE_SETUP; |
3ed75eb8f
|
254 |
vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); |
462e635e5
|
255 |
|
b6a2fea39
|
256 |
err = insert_vm_struct(mm, vma); |
eaccbfa56
|
257 |
if (err) |
b6a2fea39
|
258 |
goto err; |
b6a2fea39
|
259 260 261 |
mm->stack_vm = mm->total_vm = 1; up_write(&mm->mmap_sem); |
b6a2fea39
|
262 |
bprm->p = vma->vm_end - sizeof(void *); |
b6a2fea39
|
263 |
return 0; |
b6a2fea39
|
264 |
err: |
eaccbfa56
|
265 |
up_write(&mm->mmap_sem); |
f268dfe90
|
266 |
err_free: |
eaccbfa56
|
267 |
bprm->vma = NULL; |
3928d4f5e
|
268 |
vm_area_free(vma); |
b6a2fea39
|
269 270 271 272 273 274 275 276 277 |
return err; } static bool valid_arg_len(struct linux_binprm *bprm, long len) { return len <= MAX_ARG_STRLEN; } #else |
0e028465d
|
278 |
static inline void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) |
3c77f8457
|
279 280 |
{ } |
0e028465d
|
281 |
static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, |
b6a2fea39
|
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 |
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(). */ |
9cc64ceaa
|
341 |
static int bprm_mm_init(struct linux_binprm *bprm) |
b6a2fea39
|
342 343 344 345 346 347 348 349 |
{ int err; struct mm_struct *mm = NULL; bprm->mm = mm = mm_alloc(); err = -ENOMEM; if (!mm) goto err; |
c31dbb146
|
350 351 352 353 |
/* Save current stack limit for all calculations made during exec. */ task_lock(current->group_leader); bprm->rlim_stack = current->signal->rlim[RLIMIT_STACK]; task_unlock(current->group_leader); |
b6a2fea39
|
354 355 356 357 358 359 360 361 362 363 364 365 366 367 |
err = __bprm_mm_init(bprm); if (err) goto err; return 0; err: if (mm) { bprm->mm = NULL; mmdrop(mm); } return err; } |
ba2d01629
|
368 |
struct user_arg_ptr { |
0e028465d
|
369 370 371 372 373 374 |
#ifdef CONFIG_COMPAT bool is_compat; #endif union { const char __user *const __user *native; #ifdef CONFIG_COMPAT |
38b983b34
|
375 |
const compat_uptr_t __user *compat; |
0e028465d
|
376 377 |
#endif } ptr; |
ba2d01629
|
378 379 380 |
}; static const char __user *get_user_arg_ptr(struct user_arg_ptr argv, int nr) |
1d1dbf813
|
381 |
{ |
0e028465d
|
382 383 384 385 386 387 388 389 |
const char __user *native; #ifdef CONFIG_COMPAT if (unlikely(argv.is_compat)) { compat_uptr_t compat; if (get_user(compat, argv.ptr.compat + nr)) return ERR_PTR(-EFAULT); |
1d1dbf813
|
390 |
|
0e028465d
|
391 392 393 394 395 |
return compat_ptr(compat); } #endif if (get_user(native, argv.ptr.native + nr)) |
1d1dbf813
|
396 |
return ERR_PTR(-EFAULT); |
0e028465d
|
397 |
return native; |
1d1dbf813
|
398 |
} |
1da177e4c
|
399 400 401 |
/* * count() counts the number of strings in array ARGV. */ |
ba2d01629
|
402 |
static int count(struct user_arg_ptr argv, int max) |
1da177e4c
|
403 404 |
{ int i = 0; |
0e028465d
|
405 |
if (argv.ptr.native != NULL) { |
1da177e4c
|
406 |
for (;;) { |
1d1dbf813
|
407 |
const char __user *p = get_user_arg_ptr(argv, i); |
1da177e4c
|
408 |
|
1da177e4c
|
409 410 |
if (!p) break; |
1d1dbf813
|
411 412 413 |
if (IS_ERR(p)) return -EFAULT; |
6d92d4f6a
|
414 |
if (i >= max) |
1da177e4c
|
415 |
return -E2BIG; |
6d92d4f6a
|
416 |
++i; |
9aea5a65a
|
417 418 419 |
if (fatal_signal_pending(current)) return -ERESTARTNOHAND; |
1da177e4c
|
420 421 422 423 424 |
cond_resched(); } } return i; } |
655c16a8c
|
425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 |
static int prepare_arg_pages(struct linux_binprm *bprm, struct user_arg_ptr argv, struct user_arg_ptr envp) { unsigned long limit, ptr_size; bprm->argc = count(argv, MAX_ARG_STRINGS); if (bprm->argc < 0) return bprm->argc; bprm->envc = count(envp, MAX_ARG_STRINGS); if (bprm->envc < 0) return bprm->envc; /* * Limit to 1/4 of the max stack size or 3/4 of _STK_LIM * (whichever is smaller) 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. */ limit = _STK_LIM / 4 * 3; limit = min(limit, bprm->rlim_stack.rlim_cur / 4); /* * We've historically supported up to 32 pages (ARG_MAX) * of argument strings even with small stacks */ limit = max_t(unsigned long, limit, ARG_MAX); /* * We must account for the size of all the argv and envp pointers to * the argv and envp strings, since they will also take up space in * the stack. They aren't stored until much later when we can't * signal to the parent that the child has run out of stack space. * Instead, calculate it here so it's possible to fail gracefully. */ ptr_size = (bprm->argc + bprm->envc) * sizeof(void *); if (limit <= ptr_size) return -E2BIG; limit -= ptr_size; bprm->argmin = bprm->p - limit; return 0; } |
1da177e4c
|
468 |
/* |
b6a2fea39
|
469 470 471 |
* '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
|
472 |
*/ |
ba2d01629
|
473 |
static int copy_strings(int argc, struct user_arg_ptr argv, |
75c96f858
|
474 |
struct linux_binprm *bprm) |
1da177e4c
|
475 476 477 |
{ struct page *kmapped_page = NULL; char *kaddr = NULL; |
b6a2fea39
|
478 |
unsigned long kpos = 0; |
1da177e4c
|
479 480 481 |
int ret; while (argc-- > 0) { |
d7627467b
|
482 |
const char __user *str; |
1da177e4c
|
483 484 |
int len; unsigned long pos; |
1d1dbf813
|
485 486 487 |
ret = -EFAULT; str = get_user_arg_ptr(argv, argc); if (IS_ERR(str)) |
1da177e4c
|
488 |
goto out; |
1da177e4c
|
489 |
|
1d1dbf813
|
490 491 492 493 494 495 |
len = strnlen_user(str, MAX_ARG_STRLEN); if (!len) goto out; ret = -E2BIG; if (!valid_arg_len(bprm, len)) |
1da177e4c
|
496 |
goto out; |
1da177e4c
|
497 |
|
b6a2fea39
|
498 |
/* We're going to work our way backwords. */ |
1da177e4c
|
499 |
pos = bprm->p; |
b6a2fea39
|
500 501 |
str += len; bprm->p -= len; |
655c16a8c
|
502 503 504 505 |
#ifdef CONFIG_MMU if (bprm->p < bprm->argmin) goto out; #endif |
1da177e4c
|
506 507 |
while (len > 0) { |
1da177e4c
|
508 |
int offset, bytes_to_copy; |
1da177e4c
|
509 |
|
9aea5a65a
|
510 511 512 513 |
if (fatal_signal_pending(current)) { ret = -ERESTARTNOHAND; goto out; } |
7993bc1f4
|
514 |
cond_resched(); |
1da177e4c
|
515 |
offset = pos % PAGE_SIZE; |
b6a2fea39
|
516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 |
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
|
532 |
if (!page) { |
b6a2fea39
|
533 |
ret = -E2BIG; |
1da177e4c
|
534 535 |
goto out; } |
1da177e4c
|
536 |
|
b6a2fea39
|
537 538 |
if (kmapped_page) { flush_kernel_dcache_page(kmapped_page); |
1da177e4c
|
539 |
kunmap(kmapped_page); |
b6a2fea39
|
540 541 |
put_arg_page(kmapped_page); } |
1da177e4c
|
542 543 |
kmapped_page = page; kaddr = kmap(kmapped_page); |
b6a2fea39
|
544 545 |
kpos = pos & PAGE_MASK; flush_arg_page(bprm, kpos, kmapped_page); |
1da177e4c
|
546 |
} |
b6a2fea39
|
547 |
if (copy_from_user(kaddr+offset, str, bytes_to_copy)) { |
1da177e4c
|
548 549 550 |
ret = -EFAULT; goto out; } |
1da177e4c
|
551 552 553 554 |
} } ret = 0; out: |
b6a2fea39
|
555 556 |
if (kmapped_page) { flush_kernel_dcache_page(kmapped_page); |
1da177e4c
|
557 |
kunmap(kmapped_page); |
b6a2fea39
|
558 559 |
put_arg_page(kmapped_page); } |
1da177e4c
|
560 561 562 563 564 565 |
return ret; } /* * Like copy_strings, but get argv and its values from kernel memory. */ |
ba2d01629
|
566 |
int copy_strings_kernel(int argc, const char *const *__argv, |
d7627467b
|
567 |
struct linux_binprm *bprm) |
1da177e4c
|
568 569 570 |
{ int r; mm_segment_t oldfs = get_fs(); |
ba2d01629
|
571 |
struct user_arg_ptr argv = { |
0e028465d
|
572 |
.ptr.native = (const char __user *const __user *)__argv, |
ba2d01629
|
573 |
}; |
1da177e4c
|
574 |
set_fs(KERNEL_DS); |
ba2d01629
|
575 |
r = copy_strings(argc, argv, bprm); |
1da177e4c
|
576 |
set_fs(oldfs); |
ba2d01629
|
577 |
|
1da177e4c
|
578 579 |
return r; } |
1da177e4c
|
580 581 582 |
EXPORT_SYMBOL(copy_strings_kernel); #ifdef CONFIG_MMU |
b6a2fea39
|
583 |
|
1da177e4c
|
584 |
/* |
b6a2fea39
|
585 586 587 |
* 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
|
588 |
* |
b6a2fea39
|
589 590 591 592 593 594 |
* 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
|
595 |
*/ |
b6a2fea39
|
596 |
static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) |
1da177e4c
|
597 598 |
{ struct mm_struct *mm = vma->vm_mm; |
b6a2fea39
|
599 600 601 602 603 |
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; |
d16dfc550
|
604 |
struct mmu_gather tlb; |
1da177e4c
|
605 |
|
b6a2fea39
|
606 |
BUG_ON(new_start > new_end); |
1da177e4c
|
607 |
|
b6a2fea39
|
608 609 610 611 612 613 614 615 616 617 |
/* * 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
|
618 619 |
if (vma_adjust(vma, new_start, old_end, vma->vm_pgoff, NULL)) return -ENOMEM; |
b6a2fea39
|
620 621 622 623 624 625 |
/* * 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, |
38a76013a
|
626 |
vma, new_start, length, false)) |
b6a2fea39
|
627 628 629 |
return -ENOMEM; lru_add_drain(); |
2b047252d
|
630 |
tlb_gather_mmu(&tlb, mm, old_start, old_end); |
b6a2fea39
|
631 632 633 634 |
if (new_end > old_start) { /* * when the old and new regions overlap clear from new_end. */ |
d16dfc550
|
635 |
free_pgd_range(&tlb, new_end, old_end, new_end, |
6ee8630e0
|
636 |
vma->vm_next ? vma->vm_next->vm_start : USER_PGTABLES_CEILING); |
b6a2fea39
|
637 638 639 640 641 642 643 |
} 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. */ |
d16dfc550
|
644 |
free_pgd_range(&tlb, old_start, old_end, new_end, |
6ee8630e0
|
645 |
vma->vm_next ? vma->vm_next->vm_start : USER_PGTABLES_CEILING); |
1da177e4c
|
646 |
} |
2b047252d
|
647 |
tlb_finish_mmu(&tlb, old_start, old_end); |
b6a2fea39
|
648 649 |
/* |
5beb49305
|
650 |
* Shrink the vma to just the new range. Always succeeds. |
b6a2fea39
|
651 652 653 654 |
*/ vma_adjust(vma, new_start, new_end, vma->vm_pgoff, NULL); return 0; |
1da177e4c
|
655 |
} |
b6a2fea39
|
656 657 658 659 |
/* * Finalizes the stack vm_area_struct. The flags and permissions are updated, * the stack is optionally relocated, and some extra space is added. */ |
1da177e4c
|
660 661 662 663 |
int setup_arg_pages(struct linux_binprm *bprm, unsigned long stack_top, int executable_stack) { |
b6a2fea39
|
664 665 |
unsigned long ret; unsigned long stack_shift; |
1da177e4c
|
666 |
struct mm_struct *mm = current->mm; |
b6a2fea39
|
667 668 669 670 |
struct vm_area_struct *vma = bprm->vma; struct vm_area_struct *prev = NULL; unsigned long vm_flags; unsigned long stack_base; |
803bf5ec2
|
671 672 673 |
unsigned long stack_size; unsigned long stack_expand; unsigned long rlim_stack; |
1da177e4c
|
674 675 |
#ifdef CONFIG_STACK_GROWSUP |
d71f290b4
|
676 |
/* Limit stack size */ |
c31dbb146
|
677 |
stack_base = bprm->rlim_stack.rlim_max; |
d71f290b4
|
678 679 |
if (stack_base > STACK_SIZE_MAX) stack_base = STACK_SIZE_MAX; |
1da177e4c
|
680 |
|
d045c77c1
|
681 682 |
/* Add space for stack randomization. */ stack_base += (STACK_RND_MASK << PAGE_SHIFT); |
b6a2fea39
|
683 684 685 |
/* Make sure we didn't let the argument array grow too large. */ if (vma->vm_end - vma->vm_start > stack_base) return -ENOMEM; |
1da177e4c
|
686 |
|
b6a2fea39
|
687 |
stack_base = PAGE_ALIGN(stack_top - stack_base); |
1da177e4c
|
688 |
|
b6a2fea39
|
689 690 691 |
stack_shift = vma->vm_start - stack_base; mm->arg_start = bprm->p - stack_shift; bprm->p = vma->vm_end - stack_shift; |
1da177e4c
|
692 |
#else |
b6a2fea39
|
693 694 |
stack_top = arch_align_stack(stack_top); stack_top = PAGE_ALIGN(stack_top); |
1b528181b
|
695 696 697 698 |
if (unlikely(stack_top < mmap_min_addr) || unlikely(vma->vm_end - vma->vm_start >= stack_top - mmap_min_addr)) return -ENOMEM; |
b6a2fea39
|
699 700 701 |
stack_shift = vma->vm_end - stack_top; bprm->p -= stack_shift; |
1da177e4c
|
702 |
mm->arg_start = bprm->p; |
1da177e4c
|
703 |
#endif |
1da177e4c
|
704 |
if (bprm->loader) |
b6a2fea39
|
705 706 |
bprm->loader -= stack_shift; bprm->exec -= stack_shift; |
1da177e4c
|
707 |
|
f268dfe90
|
708 709 |
if (down_write_killable(&mm->mmap_sem)) return -EINTR; |
96a8e13ed
|
710 |
vm_flags = VM_STACK_FLAGS; |
b6a2fea39
|
711 712 713 714 715 716 717 718 719 720 721 |
/* * 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
|
722 |
vm_flags |= VM_STACK_INCOMPLETE_SETUP; |
b6a2fea39
|
723 724 725 726 727 728 |
ret = mprotect_fixup(vma, &prev, vma->vm_start, vma->vm_end, vm_flags); if (ret) goto out_unlock; BUG_ON(prev != vma); |
47a2ebb7f
|
729 730 731 732 733 |
if (unlikely(vm_flags & VM_EXEC)) { pr_warn_once("process '%pD4' started with executable stack ", bprm->file); } |
b6a2fea39
|
734 735 736 |
/* Move stack pages down in memory. */ if (stack_shift) { ret = shift_arg_pages(vma, stack_shift); |
fc63cf237
|
737 738 |
if (ret) goto out_unlock; |
1da177e4c
|
739 |
} |
a8bef8ff6
|
740 741 |
/* mprotect_fixup is overkill to remove the temporary stack flags */ vma->vm_flags &= ~VM_STACK_INCOMPLETE_SETUP; |
5ef097dd7
|
742 |
stack_expand = 131072UL; /* randomly 32*4k (or 2*64k) pages */ |
803bf5ec2
|
743 744 745 746 747 |
stack_size = vma->vm_end - vma->vm_start; /* * Align this down to a page boundary as expand_stack * will align it up. */ |
c31dbb146
|
748 |
rlim_stack = bprm->rlim_stack.rlim_cur & PAGE_MASK; |
b6a2fea39
|
749 |
#ifdef CONFIG_STACK_GROWSUP |
803bf5ec2
|
750 751 752 753 |
if (stack_size + stack_expand > rlim_stack) stack_base = vma->vm_start + rlim_stack; else stack_base = vma->vm_end + stack_expand; |
b6a2fea39
|
754 |
#else |
803bf5ec2
|
755 756 757 758 |
if (stack_size + stack_expand > rlim_stack) stack_base = vma->vm_end - rlim_stack; else stack_base = vma->vm_start - stack_expand; |
b6a2fea39
|
759 |
#endif |
3af9e8592
|
760 |
current->mm->start_stack = bprm->p; |
b6a2fea39
|
761 762 763 764 765 |
ret = expand_stack(vma, stack_base); if (ret) ret = -EFAULT; out_unlock: |
1da177e4c
|
766 |
up_write(&mm->mmap_sem); |
fc63cf237
|
767 |
return ret; |
1da177e4c
|
768 |
} |
1da177e4c
|
769 |
EXPORT_SYMBOL(setup_arg_pages); |
7e7ec6a93
|
770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 |
#else /* * Transfer the program arguments and environment from the holding pages * onto the stack. The provided stack pointer is adjusted accordingly. */ int transfer_args_to_stack(struct linux_binprm *bprm, unsigned long *sp_location) { unsigned long index, stop, sp; int ret = 0; stop = bprm->p >> PAGE_SHIFT; sp = *sp_location; for (index = MAX_ARG_PAGES - 1; index >= stop; index--) { unsigned int offset = index == stop ? bprm->p & ~PAGE_MASK : 0; char *src = kmap(bprm->page[index]) + offset; sp -= PAGE_SIZE - offset; if (copy_to_user((void *) sp, src, PAGE_SIZE - offset) != 0) ret = -EFAULT; kunmap(bprm->page[index]); if (ret) goto out; } *sp_location = sp; out: return ret; } EXPORT_SYMBOL(transfer_args_to_stack); |
1da177e4c
|
802 |
#endif /* CONFIG_MMU */ |
51f39a1f0
|
803 |
static struct file *do_open_execat(int fd, struct filename *name, int flags) |
1da177e4c
|
804 |
{ |
1da177e4c
|
805 |
struct file *file; |
e56b6a5dd
|
806 |
int err; |
51f39a1f0
|
807 |
struct open_flags open_exec_flags = { |
47c805dc2
|
808 |
.open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, |
62fb4a155
|
809 |
.acc_mode = MAY_EXEC, |
f9652e10c
|
810 811 |
.intent = LOOKUP_OPEN, .lookup_flags = LOOKUP_FOLLOW, |
47c805dc2
|
812 |
}; |
1da177e4c
|
813 |
|
51f39a1f0
|
814 815 816 817 818 819 820 821 |
if ((flags & ~(AT_SYMLINK_NOFOLLOW | AT_EMPTY_PATH)) != 0) return ERR_PTR(-EINVAL); if (flags & AT_SYMLINK_NOFOLLOW) open_exec_flags.lookup_flags &= ~LOOKUP_FOLLOW; if (flags & AT_EMPTY_PATH) open_exec_flags.lookup_flags |= LOOKUP_EMPTY; file = do_filp_open(fd, name, &open_exec_flags); |
6e8341a11
|
822 |
if (IS_ERR(file)) |
e56b6a5dd
|
823 824 825 |
goto out; err = -EACCES; |
496ad9aa8
|
826 |
if (!S_ISREG(file_inode(file)->i_mode)) |
6e8341a11
|
827 |
goto exit; |
e56b6a5dd
|
828 |
|
90f8572b0
|
829 |
if (path_noexec(&file->f_path)) |
6e8341a11
|
830 |
goto exit; |
e56b6a5dd
|
831 832 |
err = deny_write_access(file); |
6e8341a11
|
833 834 |
if (err) goto exit; |
1da177e4c
|
835 |
|
51f39a1f0
|
836 837 |
if (name->name[0] != '\0') fsnotify_open(file); |
6e8341a11
|
838 |
out: |
e56b6a5dd
|
839 |
return file; |
6e8341a11
|
840 841 |
exit: fput(file); |
e56b6a5dd
|
842 843 |
return ERR_PTR(err); } |
c4ad8f98b
|
844 845 846 |
struct file *open_exec(const char *name) { |
516891041
|
847 848 849 850 851 852 853 854 |
struct filename *filename = getname_kernel(name); struct file *f = ERR_CAST(filename); if (!IS_ERR(filename)) { f = do_open_execat(AT_FDCWD, filename, 0); putname(filename); } return f; |
c4ad8f98b
|
855 |
} |
1da177e4c
|
856 |
EXPORT_SYMBOL(open_exec); |
b44a7dfc6
|
857 |
int kernel_read_file(struct file *file, void **buf, loff_t *size, |
bc8ca5b92
|
858 |
loff_t max_size, enum kernel_read_file_id id) |
b44a7dfc6
|
859 860 861 862 863 864 865 |
{ loff_t i_size, pos; ssize_t bytes = 0; int ret; if (!S_ISREG(file_inode(file)->i_mode) || max_size < 0) return -EINVAL; |
7bd698b3c
|
866 |
ret = deny_write_access(file); |
39eeb4fb9
|
867 868 |
if (ret) return ret; |
7bd698b3c
|
869 |
ret = security_kernel_read_file(file, id); |
39d637af5
|
870 |
if (ret) |
7bd698b3c
|
871 |
goto out; |
39d637af5
|
872 |
|
b44a7dfc6
|
873 |
i_size = i_size_read(file_inode(file)); |
39d637af5
|
874 875 876 877 |
if (i_size <= 0) { ret = -EINVAL; goto out; } |
691115c35
|
878 879 880 881 |
if (i_size > SIZE_MAX || (max_size > 0 && i_size > max_size)) { ret = -EFBIG; goto out; } |
b44a7dfc6
|
882 |
|
a098ecd2f
|
883 884 |
if (id != READING_FIRMWARE_PREALLOC_BUFFER) *buf = vmalloc(i_size); |
39d637af5
|
885 886 887 888 |
if (!*buf) { ret = -ENOMEM; goto out; } |
b44a7dfc6
|
889 890 891 |
pos = 0; while (pos < i_size) { |
bdd1d2d3d
|
892 |
bytes = kernel_read(file, *buf + pos, i_size - pos, &pos); |
b44a7dfc6
|
893 894 |
if (bytes < 0) { ret = bytes; |
f612acfae
|
895 |
goto out_free; |
b44a7dfc6
|
896 897 898 899 |
} if (bytes == 0) break; |
b44a7dfc6
|
900 901 902 903 |
} if (pos != i_size) { ret = -EIO; |
39d637af5
|
904 |
goto out_free; |
b44a7dfc6
|
905 |
} |
bc8ca5b92
|
906 |
ret = security_kernel_post_read_file(file, *buf, i_size, id); |
b44a7dfc6
|
907 908 |
if (!ret) *size = pos; |
39d637af5
|
909 |
out_free: |
b44a7dfc6
|
910 |
if (ret < 0) { |
a098ecd2f
|
911 912 913 914 |
if (id != READING_FIRMWARE_PREALLOC_BUFFER) { vfree(*buf); *buf = NULL; } |
b44a7dfc6
|
915 |
} |
39d637af5
|
916 917 918 |
out: allow_write_access(file); |
b44a7dfc6
|
919 920 921 |
return ret; } EXPORT_SYMBOL_GPL(kernel_read_file); |
711aab1db
|
922 |
int kernel_read_file_from_path(const char *path, void **buf, loff_t *size, |
09596b94f
|
923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 |
loff_t max_size, enum kernel_read_file_id id) { struct file *file; int ret; if (!path || !*path) return -EINVAL; file = filp_open(path, O_RDONLY, 0); if (IS_ERR(file)) return PTR_ERR(file); ret = kernel_read_file(file, buf, size, max_size, id); fput(file); return ret; } EXPORT_SYMBOL_GPL(kernel_read_file_from_path); |
b844f0ecb
|
940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 |
int kernel_read_file_from_fd(int fd, void **buf, loff_t *size, loff_t max_size, enum kernel_read_file_id id) { struct fd f = fdget(fd); int ret = -EBADF; if (!f.file) goto out; ret = kernel_read_file(f.file, buf, size, max_size, id); out: fdput(f); return ret; } EXPORT_SYMBOL_GPL(kernel_read_file_from_fd); |
3dc20cb28
|
955 956 |
ssize_t read_code(struct file *file, unsigned long addr, loff_t pos, size_t len) { |
ec6955798
|
957 |
ssize_t res = vfs_read(file, (void __user *)addr, len, &pos); |
3dc20cb28
|
958 959 960 961 962 |
if (res > 0) flush_icache_range(addr, addr + len); return res; } EXPORT_SYMBOL(read_code); |
1da177e4c
|
963 964 965 |
static int exec_mmap(struct mm_struct *mm) { struct task_struct *tsk; |
615d6e875
|
966 |
struct mm_struct *old_mm, *active_mm; |
1da177e4c
|
967 968 969 970 |
/* Notify parent that we're no longer interested in the old VM */ tsk = current; old_mm = current->mm; |
4610ba7ad
|
971 |
exec_mm_release(tsk, old_mm); |
1da177e4c
|
972 973 |
if (old_mm) { |
4fe7efdbd
|
974 |
sync_mm_rss(old_mm); |
1da177e4c
|
975 976 977 978 |
/* * 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
|
979 |
* checking core_state and changing tsk->mm. |
1da177e4c
|
980 981 |
*/ down_read(&old_mm->mmap_sem); |
999d9fc16
|
982 |
if (unlikely(old_mm->core_state)) { |
1da177e4c
|
983 984 985 986 987 988 |
up_read(&old_mm->mmap_sem); return -EINTR; } } task_lock(tsk); active_mm = tsk->active_mm; |
227a4aadc
|
989 |
membarrier_exec_mmap(mm); |
1da177e4c
|
990 991 992 |
tsk->mm = mm; tsk->active_mm = mm; activate_mm(active_mm, mm); |
615d6e875
|
993 994 |
tsk->mm->vmacache_seqnum = 0; vmacache_flush(tsk); |
1da177e4c
|
995 |
task_unlock(tsk); |
1da177e4c
|
996 997 |
if (old_mm) { up_read(&old_mm->mmap_sem); |
7dddb12c6
|
998 |
BUG_ON(active_mm != old_mm); |
701085b21
|
999 |
setmax_mm_hiwater_rss(&tsk->signal->maxrss, old_mm); |
31a78f23b
|
1000 |
mm_update_next_owner(old_mm); |
1da177e4c
|
1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 |
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
|
1014 |
static int de_thread(struct task_struct *tsk) |
1da177e4c
|
1015 1016 |
{ struct signal_struct *sig = tsk->signal; |
b2c903b87
|
1017 |
struct sighand_struct *oldsighand = tsk->sighand; |
1da177e4c
|
1018 |
spinlock_t *lock = &oldsighand->siglock; |
1da177e4c
|
1019 |
|
aafe6c2a2
|
1020 |
if (thread_group_empty(tsk)) |
1da177e4c
|
1021 1022 1023 1024 |
goto no_thread_group; /* * Kill all other threads in the thread group. |
1da177e4c
|
1025 |
*/ |
1da177e4c
|
1026 |
spin_lock_irq(lock); |
ed5d2cac1
|
1027 |
if (signal_group_exit(sig)) { |
1da177e4c
|
1028 1029 1030 1031 1032 |
/* * Another group action in progress, just * return so that the signal is processed. */ spin_unlock_irq(lock); |
1da177e4c
|
1033 1034 |
return -EAGAIN; } |
d344193a0
|
1035 |
|
ed5d2cac1
|
1036 |
sig->group_exit_task = tsk; |
d344193a0
|
1037 1038 1039 |
sig->notify_count = zap_other_threads(tsk); if (!thread_group_leader(tsk)) sig->notify_count--; |
1da177e4c
|
1040 |
|
d344193a0
|
1041 |
while (sig->notify_count) { |
d5bbd43d5
|
1042 |
__set_current_state(TASK_KILLABLE); |
1da177e4c
|
1043 |
spin_unlock_irq(lock); |
a72173ecf
|
1044 |
schedule(); |
08d405c8b
|
1045 |
if (__fatal_signal_pending(tsk)) |
d5bbd43d5
|
1046 |
goto killed; |
1da177e4c
|
1047 1048 |
spin_lock_irq(lock); } |
1da177e4c
|
1049 1050 1051 1052 1053 1054 1055 |
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
|
1056 |
if (!thread_group_leader(tsk)) { |
8187926bd
|
1057 |
struct task_struct *leader = tsk->group_leader; |
6db840fa7
|
1058 |
|
6db840fa7
|
1059 |
for (;;) { |
780de9dd2
|
1060 |
cgroup_threadgroup_change_begin(tsk); |
6db840fa7
|
1061 |
write_lock_irq(&tasklist_lock); |
dfcce791f
|
1062 1063 1064 1065 1066 |
/* * Do this under tasklist_lock to ensure that * exit_notify() can't miss ->group_exit_task */ sig->notify_count = -1; |
6db840fa7
|
1067 1068 |
if (likely(leader->exit_state)) break; |
d5bbd43d5
|
1069 |
__set_current_state(TASK_KILLABLE); |
6db840fa7
|
1070 |
write_unlock_irq(&tasklist_lock); |
780de9dd2
|
1071 |
cgroup_threadgroup_change_end(tsk); |
a72173ecf
|
1072 |
schedule(); |
08d405c8b
|
1073 |
if (__fatal_signal_pending(tsk)) |
d5bbd43d5
|
1074 |
goto killed; |
6db840fa7
|
1075 |
} |
1da177e4c
|
1076 |
|
f5e902817
|
1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 |
/* * 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
|
1087 |
tsk->start_time = leader->start_time; |
cf25e24db
|
1088 |
tsk->start_boottime = leader->start_boottime; |
f5e902817
|
1089 |
|
bac0abd61
|
1090 1091 |
BUG_ON(!same_thread_group(leader, tsk)); BUG_ON(has_group_leader_pid(tsk)); |
1da177e4c
|
1092 1093 1094 1095 1096 1097 |
/* * 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
|
1098 1099 |
/* Become a process group leader with the old leader's pid. |
c18258c6f
|
1100 1101 |
* The old leader becomes a thread of the this thread group. * Note: The old leader also uses this pid until release_task |
d73d65293
|
1102 1103 |
* is called. Odd but simple and correct. */ |
aafe6c2a2
|
1104 |
tsk->pid = leader->pid; |
3f4185483
|
1105 |
change_pid(tsk, PIDTYPE_PID, task_pid(leader)); |
6883f81aa
|
1106 |
transfer_pid(leader, tsk, PIDTYPE_TGID); |
aafe6c2a2
|
1107 1108 |
transfer_pid(leader, tsk, PIDTYPE_PGID); transfer_pid(leader, tsk, PIDTYPE_SID); |
9cd80bbb0
|
1109 |
|
aafe6c2a2
|
1110 |
list_replace_rcu(&leader->tasks, &tsk->tasks); |
9cd80bbb0
|
1111 |
list_replace_init(&leader->sibling, &tsk->sibling); |
1da177e4c
|
1112 |
|
aafe6c2a2
|
1113 1114 |
tsk->group_leader = tsk; leader->group_leader = tsk; |
de12a7878
|
1115 |
|
aafe6c2a2
|
1116 |
tsk->exit_signal = SIGCHLD; |
087806b12
|
1117 |
leader->exit_signal = -1; |
962b564cf
|
1118 1119 1120 |
BUG_ON(leader->exit_state != EXIT_ZOMBIE); leader->exit_state = EXIT_DEAD; |
eac1b5e57
|
1121 1122 1123 1124 1125 1126 1127 1128 |
/* * We are going to release_task()->ptrace_unlink() silently, * the tracer can sleep in do_wait(). EXIT_DEAD guarantees * the tracer wont't block again waiting for this thread. */ if (unlikely(leader->ptrace)) __wake_up_parent(leader, leader->parent); |
1da177e4c
|
1129 |
write_unlock_irq(&tasklist_lock); |
780de9dd2
|
1130 |
cgroup_threadgroup_change_end(tsk); |
8187926bd
|
1131 1132 |
release_task(leader); |
ed5d2cac1
|
1133 |
} |
1da177e4c
|
1134 |
|
6db840fa7
|
1135 1136 |
sig->group_exit_task = NULL; sig->notify_count = 0; |
1da177e4c
|
1137 1138 |
no_thread_group: |
e63682534
|
1139 1140 |
/* we have changed execution domain */ tsk->exit_signal = SIGCHLD; |
baa73d9e4
|
1141 |
#ifdef CONFIG_POSIX_TIMERS |
1da177e4c
|
1142 |
exit_itimers(sig); |
cbaffba12
|
1143 |
flush_itimer_signals(); |
baa73d9e4
|
1144 |
#endif |
329f7dba5
|
1145 |
|
d036bda7d
|
1146 |
if (refcount_read(&oldsighand->count) != 1) { |
b2c903b87
|
1147 |
struct sighand_struct *newsighand; |
1da177e4c
|
1148 |
/* |
b2c903b87
|
1149 1150 |
* This ->sighand is shared with the CLONE_SIGHAND * but not CLONE_THREAD task, switch to the new one. |
1da177e4c
|
1151 |
*/ |
b2c903b87
|
1152 1153 1154 |
newsighand = kmem_cache_alloc(sighand_cachep, GFP_KERNEL); if (!newsighand) return -ENOMEM; |
d036bda7d
|
1155 |
refcount_set(&newsighand->count, 1); |
1da177e4c
|
1156 1157 1158 1159 1160 |
memcpy(newsighand->action, oldsighand->action, sizeof(newsighand->action)); write_lock_irq(&tasklist_lock); spin_lock(&oldsighand->siglock); |
aafe6c2a2
|
1161 |
rcu_assign_pointer(tsk->sighand, newsighand); |
1da177e4c
|
1162 1163 |
spin_unlock(&oldsighand->siglock); write_unlock_irq(&tasklist_lock); |
fba2afaae
|
1164 |
__cleanup_sighand(oldsighand); |
1da177e4c
|
1165 |
} |
aafe6c2a2
|
1166 |
BUG_ON(!thread_group_leader(tsk)); |
1da177e4c
|
1167 |
return 0; |
d5bbd43d5
|
1168 1169 1170 1171 1172 1173 1174 1175 |
killed: /* protects against exit_notify() and __exit_signal() */ read_lock(&tasklist_lock); sig->group_exit_task = NULL; sig->notify_count = 0; read_unlock(&tasklist_lock); return -EAGAIN; |
1da177e4c
|
1176 |
} |
0840a90d9
|
1177 |
|
3756f6401
|
1178 |
char *__get_task_comm(char *buf, size_t buf_size, struct task_struct *tsk) |
1da177e4c
|
1179 |
{ |
1da177e4c
|
1180 |
task_lock(tsk); |
3756f6401
|
1181 |
strncpy(buf, tsk->comm, buf_size); |
1da177e4c
|
1182 |
task_unlock(tsk); |
59714d65d
|
1183 |
return buf; |
1da177e4c
|
1184 |
} |
3756f6401
|
1185 |
EXPORT_SYMBOL_GPL(__get_task_comm); |
1da177e4c
|
1186 |
|
6a6d27de3
|
1187 1188 1189 1190 |
/* * These functions flushes out all traces of the currently running executable * so that a new one can be started */ |
82b897782
|
1191 |
void __set_task_comm(struct task_struct *tsk, const char *buf, bool exec) |
1da177e4c
|
1192 1193 |
{ task_lock(tsk); |
43d2b1132
|
1194 |
trace_task_rename(tsk, buf); |
1da177e4c
|
1195 1196 |
strlcpy(tsk->comm, buf, sizeof(tsk->comm)); task_unlock(tsk); |
82b897782
|
1197 |
perf_event_comm(tsk, exec); |
1da177e4c
|
1198 |
} |
a9208e42b
|
1199 1200 1201 1202 1203 1204 |
/* * Calling this is the point of no return. None of the failures will be * seen by userspace since either the process is already taking a fatal * signal (via de_thread() or coredump), or will have SEGV raised * (after exec_mmap()) by search_binary_handlers (see below). */ |
1da177e4c
|
1205 1206 |
int flush_old_exec(struct linux_binprm * bprm) { |
221af7f87
|
1207 |
int retval; |
1da177e4c
|
1208 1209 1210 1211 1212 1213 1214 1215 |
/* * 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; |
6e399cd14
|
1216 1217 1218 1219 1220 |
/* * Must be called _before_ exec_mmap() as bprm->mm is * not visibile until then. This also enables the update * to be lockless. */ |
925d1c401
|
1221 |
set_mm_exe_file(bprm->mm, bprm->file); |
6e399cd14
|
1222 |
|
1da177e4c
|
1223 |
/* |
1da177e4c
|
1224 1225 |
* Release all of the old mmap stuff */ |
3c77f8457
|
1226 |
acct_arg_size(bprm, 0); |
1da177e4c
|
1227 1228 |
retval = exec_mmap(bprm->mm); if (retval) |
fd8328be8
|
1229 |
goto out; |
1da177e4c
|
1230 |
|
a9208e42b
|
1231 1232 1233 1234 1235 1236 1237 |
/* * After clearing bprm->mm (to mark that current is using the * prepared mm now), we have nothing left of the original * process. If anything from here on returns an error, the check * in search_binary_handler() will SEGV current. */ bprm->mm = NULL; |
7ab02af42
|
1238 |
|
dac853ae8
|
1239 |
set_fs(USER_DS); |
b88fae644
|
1240 1241 |
current->flags &= ~(PF_RANDOMIZE | PF_FORKNOEXEC | PF_KTHREAD | PF_NOFREEZE | PF_NO_SETAFFINITY); |
7ab02af42
|
1242 1243 |
flush_thread(); current->personality &= ~bprm->per_clear; |
613cc2b6f
|
1244 1245 1246 1247 1248 1249 1250 |
/* * We have to apply CLOEXEC before we change whether the process is * dumpable (in setup_new_exec) to avoid a race with a process in userspace * trying to access the should-be-closed file descriptors of a process * undergoing exec(2). */ do_close_on_exec(current->files); |
221af7f87
|
1251 1252 1253 1254 1255 1256 |
return 0; out: return retval; } EXPORT_SYMBOL(flush_old_exec); |
1b5d783c9
|
1257 1258 |
void would_dump(struct linux_binprm *bprm, struct file *file) { |
f84df2a6f
|
1259 1260 1261 |
struct inode *inode = file_inode(file); if (inode_permission(inode, MAY_READ) < 0) { struct user_namespace *old, *user_ns; |
1b5d783c9
|
1262 |
bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP; |
f84df2a6f
|
1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 |
/* Ensure mm->user_ns contains the executable */ user_ns = old = bprm->mm->user_ns; while ((user_ns != &init_user_ns) && !privileged_wrt_inode_uidgid(user_ns, inode)) user_ns = user_ns->parent; if (old != user_ns) { bprm->mm->user_ns = get_user_ns(user_ns); put_user_ns(old); } } |
1b5d783c9
|
1275 1276 |
} EXPORT_SYMBOL(would_dump); |
221af7f87
|
1277 1278 |
void setup_new_exec(struct linux_binprm * bprm) { |
46d98eb4e
|
1279 1280 1281 1282 1283 1284 |
/* * Once here, prepare_binrpm() will not be called any more, so * the final state of setuid/setgid/fscaps can be merged into the * secureexec flag. */ bprm->secureexec |= bprm->cap_elevated; |
64701dee4
|
1285 |
if (bprm->secureexec) { |
fe8993b3a
|
1286 1287 |
/* Make sure parent cannot signal privileged process. */ current->pdeath_signal = 0; |
64701dee4
|
1288 1289 1290 1291 1292 |
/* * For secureexec, reset the stack limit to sane default to * avoid bad behavior from the prior rlimits. This has to * happen before arch_pick_mmap_layout(), which examines * RLIMIT_STACK, but after the point of no return to avoid |
779f4e1c6
|
1293 |
* needing to clean up the change on failure. |
64701dee4
|
1294 |
*/ |
c31dbb146
|
1295 1296 |
if (bprm->rlim_stack.rlim_cur > _STK_LIM) bprm->rlim_stack.rlim_cur = _STK_LIM; |
64701dee4
|
1297 |
} |
c31dbb146
|
1298 |
arch_pick_mmap_layout(current->mm, &bprm->rlim_stack); |
1da177e4c
|
1299 |
|
1da177e4c
|
1300 |
current->sas_ss_sp = current->sas_ss_size = 0; |
e816c201a
|
1301 1302 1303 1304 1305 |
/* * Figure out dumpability. Note that this checking only of current * is wrong, but userspace depends on it. This should be testing * bprm->secureexec instead. */ |
473d89639
|
1306 |
if (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP || |
e816c201a
|
1307 1308 |
!(uid_eq(current_euid(), current_uid()) && gid_eq(current_egid(), current_gid()))) |
6c5d52382
|
1309 |
set_dumpable(current->mm, suid_dumpable); |
473d89639
|
1310 1311 |
else set_dumpable(current->mm, SUID_DUMP_USER); |
d6e711448
|
1312 |
|
e9ea1e7f5
|
1313 |
arch_setup_new_exec(); |
e041e328c
|
1314 |
perf_event_exec(); |
82b897782
|
1315 |
__set_task_comm(current, kbasename(bprm->filename), true); |
1da177e4c
|
1316 |
|
0551fbd29
|
1317 1318 1319 1320 1321 |
/* 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; |
1da177e4c
|
1322 1323 |
/* An exec changes our domain. We are no longer part of the thread group */ |
1da177e4c
|
1324 |
current->self_exec_id++; |
1da177e4c
|
1325 |
flush_signal_handlers(current, 0); |
1da177e4c
|
1326 |
} |
221af7f87
|
1327 |
EXPORT_SYMBOL(setup_new_exec); |
1da177e4c
|
1328 |
|
b83838313
|
1329 1330 1331 |
/* Runs immediately before start_thread() takes over. */ void finalize_exec(struct linux_binprm *bprm) { |
c31dbb146
|
1332 1333 1334 1335 |
/* Store any stack rlimit changes before starting thread. */ task_lock(current->group_leader); current->signal->rlim[RLIMIT_STACK] = bprm->rlim_stack; task_unlock(current->group_leader); |
b83838313
|
1336 1337 |
} EXPORT_SYMBOL(finalize_exec); |
a6f76f23d
|
1338 |
/* |
a2a8474c3
|
1339 1340 1341 1342 1343 |
* 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. */ |
4addd2640
|
1344 |
static int prepare_bprm_creds(struct linux_binprm *bprm) |
a2a8474c3
|
1345 |
{ |
9b1bf12d5
|
1346 |
if (mutex_lock_interruptible(¤t->signal->cred_guard_mutex)) |
a2a8474c3
|
1347 1348 1349 1350 1351 |
return -ERESTARTNOINTR; bprm->cred = prepare_exec_creds(); if (likely(bprm->cred)) return 0; |
9b1bf12d5
|
1352 |
mutex_unlock(¤t->signal->cred_guard_mutex); |
a2a8474c3
|
1353 1354 |
return -ENOMEM; } |
c4ad8f98b
|
1355 |
static void free_bprm(struct linux_binprm *bprm) |
a2a8474c3
|
1356 1357 1358 |
{ free_arg_pages(bprm); if (bprm->cred) { |
9b1bf12d5
|
1359 |
mutex_unlock(¤t->signal->cred_guard_mutex); |
a2a8474c3
|
1360 1361 |
abort_creds(bprm->cred); } |
63e46b95e
|
1362 1363 1364 1365 |
if (bprm->file) { allow_write_access(bprm->file); fput(bprm->file); } |
b66c59840
|
1366 1367 1368 |
/* If a binfmt changed the interp, free it. */ if (bprm->interp != bprm->filename) kfree(bprm->interp); |
a2a8474c3
|
1369 1370 |
kfree(bprm); } |
c2315c187
|
1371 |
int bprm_change_interp(const char *interp, struct linux_binprm *bprm) |
b66c59840
|
1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 |
{ /* If a binfmt changed the interp, free it first. */ if (bprm->interp != bprm->filename) kfree(bprm->interp); bprm->interp = kstrdup(interp, GFP_KERNEL); if (!bprm->interp) return -ENOMEM; return 0; } EXPORT_SYMBOL(bprm_change_interp); |
a2a8474c3
|
1382 |
/* |
a6f76f23d
|
1383 1384 1385 1386 1387 1388 1389 1390 |
* 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; |
2976b10f0
|
1391 1392 1393 1394 1395 1396 1397 1398 1399 |
/* * Disable monitoring for regular users * when executing setuid binaries. Must * wait until new credentials are committed * by commit_creds() above */ if (get_dumpable(current->mm) != SUID_DUMP_USER) perf_event_exit_task(current); |
a2a8474c3
|
1400 1401 |
/* * cred_guard_mutex must be held at least to this point to prevent |
a6f76f23d
|
1402 |
* ptrace_attach() from altering our determination of the task's |
a2a8474c3
|
1403 1404 |
* credentials; any time after this it may be unlocked. */ |
a6f76f23d
|
1405 |
security_bprm_committed_creds(bprm); |
9b1bf12d5
|
1406 |
mutex_unlock(¤t->signal->cred_guard_mutex); |
a6f76f23d
|
1407 1408 1409 1410 1411 |
} EXPORT_SYMBOL(install_exec_creds); /* * determine how safe it is to execute the proposed program |
9b1bf12d5
|
1412 |
* - the caller must hold ->cred_guard_mutex to protect against |
c2e1f2e30
|
1413 |
* PTRACE_ATTACH or seccomp thread-sync |
a6f76f23d
|
1414 |
*/ |
9e00cdb09
|
1415 |
static void check_unsafe_exec(struct linux_binprm *bprm) |
a6f76f23d
|
1416 |
{ |
0bf2f3aec
|
1417 |
struct task_struct *p = current, *t; |
f1191b50e
|
1418 |
unsigned n_fs; |
a6f76f23d
|
1419 |
|
9227dd2a8
|
1420 1421 |
if (p->ptrace) bprm->unsafe |= LSM_UNSAFE_PTRACE; |
a6f76f23d
|
1422 |
|
259e5e6c7
|
1423 1424 1425 1426 |
/* * This isn't strictly necessary, but it makes it harder for LSMs to * mess up. */ |
1d4457f99
|
1427 |
if (task_no_new_privs(current)) |
259e5e6c7
|
1428 |
bprm->unsafe |= LSM_UNSAFE_NO_NEW_PRIVS; |
83f62a2ea
|
1429 |
t = p; |
0bf2f3aec
|
1430 |
n_fs = 1; |
2a4419b5b
|
1431 |
spin_lock(&p->fs->lock); |
437f7fdb6
|
1432 |
rcu_read_lock(); |
83f62a2ea
|
1433 |
while_each_thread(p, t) { |
0bf2f3aec
|
1434 1435 |
if (t->fs == p->fs) n_fs++; |
0bf2f3aec
|
1436 |
} |
437f7fdb6
|
1437 |
rcu_read_unlock(); |
0bf2f3aec
|
1438 |
|
9e00cdb09
|
1439 |
if (p->fs->users > n_fs) |
a6f76f23d
|
1440 |
bprm->unsafe |= LSM_UNSAFE_SHARE; |
9e00cdb09
|
1441 1442 |
else p->fs->in_exec = 1; |
2a4419b5b
|
1443 |
spin_unlock(&p->fs->lock); |
a6f76f23d
|
1444 |
} |
8b01fc86b
|
1445 1446 1447 1448 1449 1450 |
static void bprm_fill_uid(struct linux_binprm *bprm) { struct inode *inode; unsigned int mode; kuid_t uid; kgid_t gid; |
cb6fd68fd
|
1451 1452 1453 1454 1455 1456 |
/* * Since this can be called multiple times (via prepare_binprm), * we must clear any previous work done when setting set[ug]id * bits from any earlier bprm->file uses (for example when run * first for a setuid script then again for its interpreter). */ |
8b01fc86b
|
1457 1458 |
bprm->cred->euid = current_euid(); bprm->cred->egid = current_egid(); |
380cf5ba6
|
1459 |
if (!mnt_may_suid(bprm->file->f_path.mnt)) |
8b01fc86b
|
1460 1461 1462 1463 |
return; if (task_no_new_privs(current)) return; |
fea6d2a61
|
1464 |
inode = bprm->file->f_path.dentry->d_inode; |
8b01fc86b
|
1465 1466 1467 1468 1469 |
mode = READ_ONCE(inode->i_mode); if (!(mode & (S_ISUID|S_ISGID))) return; /* Be careful if suid/sgid is set */ |
5955102c9
|
1470 |
inode_lock(inode); |
8b01fc86b
|
1471 1472 1473 1474 1475 |
/* reload atomically mode/uid/gid now that lock held */ mode = inode->i_mode; uid = inode->i_uid; gid = inode->i_gid; |
5955102c9
|
1476 |
inode_unlock(inode); |
8b01fc86b
|
1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 |
/* We ignore suid/sgid if there are no mappings for them in the ns */ if (!kuid_has_mapping(bprm->cred->user_ns, uid) || !kgid_has_mapping(bprm->cred->user_ns, gid)) return; if (mode & S_ISUID) { bprm->per_clear |= PER_CLEAR_ON_SETID; bprm->cred->euid = uid; } if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP)) { bprm->per_clear |= PER_CLEAR_ON_SETID; bprm->cred->egid = gid; } } |
9e00cdb09
|
1493 1494 |
/* * Fill the binprm structure from the inode. |
6eb3c3d0a
|
1495 |
* Check permissions, then read the first BINPRM_BUF_SIZE bytes |
a6f76f23d
|
1496 1497 |
* * This may be called multiple times for binary chains (scripts for example). |
1da177e4c
|
1498 1499 1500 |
*/ int prepare_binprm(struct linux_binprm *bprm) { |
1da177e4c
|
1501 |
int retval; |
bdd1d2d3d
|
1502 |
loff_t pos = 0; |
1da177e4c
|
1503 |
|
8b01fc86b
|
1504 |
bprm_fill_uid(bprm); |
1da177e4c
|
1505 1506 |
/* fill in binprm security blob */ |
a6f76f23d
|
1507 |
retval = security_bprm_set_creds(bprm); |
1da177e4c
|
1508 1509 |
if (retval) return retval; |
ddb4a1442
|
1510 |
bprm->called_set_creds = 1; |
1da177e4c
|
1511 |
|
a6f76f23d
|
1512 |
memset(bprm->buf, 0, BINPRM_BUF_SIZE); |
bdd1d2d3d
|
1513 |
return kernel_read(bprm->file, bprm->buf, BINPRM_BUF_SIZE, &pos); |
1da177e4c
|
1514 1515 1516 |
} EXPORT_SYMBOL(prepare_binprm); |
4fc75ff48
|
1517 1518 1519 1520 1521 |
/* * 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
|
1522 |
int remove_arg_zero(struct linux_binprm *bprm) |
1da177e4c
|
1523 |
{ |
b6a2fea39
|
1524 1525 1526 1527 |
int ret = 0; unsigned long offset; char *kaddr; struct page *page; |
4fc75ff48
|
1528 |
|
b6a2fea39
|
1529 1530 |
if (!bprm->argc) return 0; |
1da177e4c
|
1531 |
|
b6a2fea39
|
1532 1533 1534 1535 1536 1537 1538 |
do { offset = bprm->p & ~PAGE_MASK; page = get_arg_page(bprm, bprm->p, 0); if (!page) { ret = -EFAULT; goto out; } |
e8e3c3d66
|
1539 |
kaddr = kmap_atomic(page); |
4fc75ff48
|
1540 |
|
b6a2fea39
|
1541 1542 1543 |
for (; offset < PAGE_SIZE && kaddr[offset]; offset++, bprm->p++) ; |
4fc75ff48
|
1544 |
|
e8e3c3d66
|
1545 |
kunmap_atomic(kaddr); |
b6a2fea39
|
1546 |
put_arg_page(page); |
b6a2fea39
|
1547 |
} while (offset == PAGE_SIZE); |
4fc75ff48
|
1548 |
|
b6a2fea39
|
1549 1550 1551 |
bprm->p++; bprm->argc--; ret = 0; |
4fc75ff48
|
1552 |
|
b6a2fea39
|
1553 1554 |
out: return ret; |
1da177e4c
|
1555 |
} |
1da177e4c
|
1556 |
EXPORT_SYMBOL(remove_arg_zero); |
cb7b6b1cb
|
1557 1558 |
#define printable(c) (((c)=='\t') || ((c)==' ') || (0x20<=(c) && (c)<=0x7e)) |
1da177e4c
|
1559 1560 1561 |
/* * cycle the list of binary formats handler, until one recognizes the image */ |
3c456bfc4
|
1562 |
int search_binary_handler(struct linux_binprm *bprm) |
1da177e4c
|
1563 |
{ |
cb7b6b1cb
|
1564 |
bool need_retry = IS_ENABLED(CONFIG_MODULES); |
1da177e4c
|
1565 |
struct linux_binfmt *fmt; |
cb7b6b1cb
|
1566 |
int retval; |
1da177e4c
|
1567 |
|
d74026986
|
1568 |
/* This allows 4 levels of binfmt rewrites before failing hard. */ |
131b2f9f1
|
1569 |
if (bprm->recursion_depth > 5) |
d74026986
|
1570 |
return -ELOOP; |
1da177e4c
|
1571 1572 1573 |
retval = security_bprm_check(bprm); if (retval) return retval; |
1da177e4c
|
1574 |
retval = -ENOENT; |
cb7b6b1cb
|
1575 1576 1577 1578 1579 1580 |
retry: read_lock(&binfmt_lock); list_for_each_entry(fmt, &formats, lh) { if (!try_module_get(fmt->module)) continue; read_unlock(&binfmt_lock); |
d53ddd018
|
1581 |
|
cb7b6b1cb
|
1582 1583 |
bprm->recursion_depth++; retval = fmt->load_binary(bprm); |
d53ddd018
|
1584 |
bprm->recursion_depth--; |
19d860a14
|
1585 1586 |
read_lock(&binfmt_lock); put_binfmt(fmt); |
19d860a14
|
1587 1588 1589 |
if (retval < 0 && !bprm->mm) { /* we got to flush_old_exec() and failed after it */ read_unlock(&binfmt_lock); |
cb44c9a0a
|
1590 |
force_sigsegv(SIGSEGV); |
19d860a14
|
1591 1592 1593 1594 |
return retval; } if (retval != -ENOEXEC || !bprm->file) { read_unlock(&binfmt_lock); |
cb7b6b1cb
|
1595 |
return retval; |
1da177e4c
|
1596 |
} |
1da177e4c
|
1597 |
} |
cb7b6b1cb
|
1598 |
read_unlock(&binfmt_lock); |
19d860a14
|
1599 |
if (need_retry) { |
cb7b6b1cb
|
1600 1601 1602 |
if (printable(bprm->buf[0]) && printable(bprm->buf[1]) && printable(bprm->buf[2]) && printable(bprm->buf[3])) return retval; |
4e0621a07
|
1603 1604 |
if (request_module("binfmt-%04x", *(ushort *)(bprm->buf + 2)) < 0) return retval; |
cb7b6b1cb
|
1605 1606 1607 |
need_retry = false; goto retry; } |
1da177e4c
|
1608 1609 |
return retval; } |
1da177e4c
|
1610 |
EXPORT_SYMBOL(search_binary_handler); |
5d1baf3b6
|
1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 |
static int exec_binprm(struct linux_binprm *bprm) { pid_t old_pid, old_vpid; int ret; /* Need to fetch pid before load_binary changes it */ old_pid = current->pid; rcu_read_lock(); old_vpid = task_pid_nr_ns(current, task_active_pid_ns(current->parent)); rcu_read_unlock(); ret = search_binary_handler(bprm); if (ret >= 0) { |
3eaded86a
|
1624 |
audit_bprm(bprm); |
5d1baf3b6
|
1625 1626 |
trace_sched_process_exec(current, old_pid, bprm); ptrace_event(PTRACE_EVENT_EXEC, old_vpid); |
9beb266f2
|
1627 |
proc_exec_connector(current); |
5d1baf3b6
|
1628 1629 1630 1631 |
} return ret; } |
1da177e4c
|
1632 1633 1634 |
/* * sys_execve() executes a new program. */ |
449325b52
|
1635 1636 1637 1638 |
static int __do_execve_file(int fd, struct filename *filename, struct user_arg_ptr argv, struct user_arg_ptr envp, int flags, struct file *file) |
1da177e4c
|
1639 |
{ |
51f39a1f0
|
1640 |
char *pathbuf = NULL; |
1da177e4c
|
1641 |
struct linux_binprm *bprm; |
3b1253880
|
1642 |
struct files_struct *displaced; |
1da177e4c
|
1643 |
int retval; |
72fa59970
|
1644 |
|
c4ad8f98b
|
1645 1646 |
if (IS_ERR(filename)) return PTR_ERR(filename); |
72fa59970
|
1647 1648 1649 1650 1651 1652 1653 |
/* * We move the actual failure in case of RLIMIT_NPROC excess from * set*uid() to execve() because too many poorly written programs * don't check setuid() return code. Here we additionally recheck * whether NPROC limit is still exceeded. */ if ((current->flags & PF_NPROC_EXCEEDED) && |
bd9d43f47
|
1654 |
atomic_read(¤t_user()->processes) > rlimit(RLIMIT_NPROC)) { |
72fa59970
|
1655 1656 1657 1658 1659 1660 1661 |
retval = -EAGAIN; goto out_ret; } /* We're below the limit (still or again), so we don't want to make * further execve() calls fail. */ current->flags &= ~PF_NPROC_EXCEEDED; |
1da177e4c
|
1662 |
|
3b1253880
|
1663 |
retval = unshare_files(&displaced); |
fd8328be8
|
1664 1665 |
if (retval) goto out_ret; |
1da177e4c
|
1666 |
retval = -ENOMEM; |
11b0b5abb
|
1667 |
bprm = kzalloc(sizeof(*bprm), GFP_KERNEL); |
1da177e4c
|
1668 |
if (!bprm) |
fd8328be8
|
1669 |
goto out_files; |
1da177e4c
|
1670 |
|
a2a8474c3
|
1671 1672 |
retval = prepare_bprm_creds(bprm); if (retval) |
a6f76f23d
|
1673 |
goto out_free; |
498052bba
|
1674 |
|
9e00cdb09
|
1675 |
check_unsafe_exec(bprm); |
a2a8474c3
|
1676 |
current->in_execve = 1; |
a6f76f23d
|
1677 |
|
449325b52
|
1678 1679 |
if (!file) file = do_open_execat(fd, filename, flags); |
1da177e4c
|
1680 1681 |
retval = PTR_ERR(file); if (IS_ERR(file)) |
498052bba
|
1682 |
goto out_unmark; |
1da177e4c
|
1683 1684 |
sched_exec(); |
1da177e4c
|
1685 |
bprm->file = file; |
449325b52
|
1686 1687 1688 |
if (!filename) { bprm->filename = "none"; } else if (fd == AT_FDCWD || filename->name[0] == '/') { |
51f39a1f0
|
1689 1690 1691 |
bprm->filename = filename->name; } else { if (filename->name[0] == '\0') |
0ee931c4e
|
1692 |
pathbuf = kasprintf(GFP_KERNEL, "/dev/fd/%d", fd); |
51f39a1f0
|
1693 |
else |
0ee931c4e
|
1694 |
pathbuf = kasprintf(GFP_KERNEL, "/dev/fd/%d/%s", |
51f39a1f0
|
1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 |
fd, filename->name); if (!pathbuf) { retval = -ENOMEM; goto out_unmark; } /* * Record that a name derived from an O_CLOEXEC fd will be * inaccessible after exec. Relies on having exclusive access to * current->files (due to unshare_files above). */ if (close_on_exec(fd, rcu_dereference_raw(current->files->fdt))) bprm->interp_flags |= BINPRM_FLAGS_PATH_INACCESSIBLE; bprm->filename = pathbuf; } bprm->interp = bprm->filename; |
1da177e4c
|
1710 |
|
b6a2fea39
|
1711 1712 |
retval = bprm_mm_init(bprm); if (retval) |
63e46b95e
|
1713 |
goto out_unmark; |
1da177e4c
|
1714 |
|
655c16a8c
|
1715 1716 |
retval = prepare_arg_pages(bprm, argv, envp); if (retval < 0) |
1da177e4c
|
1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 |
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; |
f84df2a6f
|
1735 |
would_dump(bprm, bprm->file); |
5d1baf3b6
|
1736 |
retval = exec_binprm(bprm); |
a6f76f23d
|
1737 1738 |
if (retval < 0) goto out; |
1da177e4c
|
1739 |
|
a6f76f23d
|
1740 |
/* execve succeeded */ |
498052bba
|
1741 |
current->fs->in_exec = 0; |
f9ce1f1cd
|
1742 |
current->in_execve = 0; |
d7822b1e2
|
1743 |
rseq_execve(current); |
a6f76f23d
|
1744 |
acct_update_integrals(current); |
16d51a590
|
1745 |
task_numa_free(current, false); |
a6f76f23d
|
1746 |
free_bprm(bprm); |
51f39a1f0
|
1747 |
kfree(pathbuf); |
449325b52
|
1748 1749 |
if (filename) putname(filename); |
a6f76f23d
|
1750 1751 1752 |
if (displaced) put_files_struct(displaced); return retval; |
1da177e4c
|
1753 |
|
a6f76f23d
|
1754 |
out: |
3c77f8457
|
1755 1756 1757 1758 |
if (bprm->mm) { acct_arg_size(bprm, 0); mmput(bprm->mm); } |
1da177e4c
|
1759 |
|
498052bba
|
1760 |
out_unmark: |
9e00cdb09
|
1761 |
current->fs->in_exec = 0; |
f9ce1f1cd
|
1762 |
current->in_execve = 0; |
a6f76f23d
|
1763 1764 |
out_free: |
08a6fac1c
|
1765 |
free_bprm(bprm); |
51f39a1f0
|
1766 |
kfree(pathbuf); |
1da177e4c
|
1767 |
|
fd8328be8
|
1768 |
out_files: |
3b1253880
|
1769 1770 |
if (displaced) reset_files_struct(displaced); |
1da177e4c
|
1771 |
out_ret: |
449325b52
|
1772 1773 |
if (filename) putname(filename); |
1da177e4c
|
1774 1775 |
return retval; } |
449325b52
|
1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 |
static int do_execveat_common(int fd, struct filename *filename, struct user_arg_ptr argv, struct user_arg_ptr envp, int flags) { return __do_execve_file(fd, filename, argv, envp, flags, NULL); } int do_execve_file(struct file *file, void *__argv, void *__envp) { struct user_arg_ptr argv = { .ptr.native = __argv }; struct user_arg_ptr envp = { .ptr.native = __envp }; return __do_execve_file(AT_FDCWD, NULL, argv, envp, 0, file); } |
c4ad8f98b
|
1791 |
int do_execve(struct filename *filename, |
ba2d01629
|
1792 |
const char __user *const __user *__argv, |
da3d4c5fa
|
1793 |
const char __user *const __user *__envp) |
ba2d01629
|
1794 |
{ |
0e028465d
|
1795 1796 |
struct user_arg_ptr argv = { .ptr.native = __argv }; struct user_arg_ptr envp = { .ptr.native = __envp }; |
51f39a1f0
|
1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 |
return do_execveat_common(AT_FDCWD, filename, argv, envp, 0); } int do_execveat(int fd, struct filename *filename, const char __user *const __user *__argv, const char __user *const __user *__envp, int flags) { struct user_arg_ptr argv = { .ptr.native = __argv }; struct user_arg_ptr envp = { .ptr.native = __envp }; return do_execveat_common(fd, filename, argv, envp, flags); |
0e028465d
|
1809 1810 1811 |
} #ifdef CONFIG_COMPAT |
c4ad8f98b
|
1812 |
static int compat_do_execve(struct filename *filename, |
38b983b34
|
1813 |
const compat_uptr_t __user *__argv, |
d03d26e58
|
1814 |
const compat_uptr_t __user *__envp) |
0e028465d
|
1815 1816 1817 1818 1819 1820 1821 1822 1823 |
{ struct user_arg_ptr argv = { .is_compat = true, .ptr.compat = __argv, }; struct user_arg_ptr envp = { .is_compat = true, .ptr.compat = __envp, }; |
51f39a1f0
|
1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 |
return do_execveat_common(AT_FDCWD, filename, argv, envp, 0); } static int compat_do_execveat(int fd, struct filename *filename, const compat_uptr_t __user *__argv, const compat_uptr_t __user *__envp, int flags) { struct user_arg_ptr argv = { .is_compat = true, .ptr.compat = __argv, }; struct user_arg_ptr envp = { .is_compat = true, .ptr.compat = __envp, }; return do_execveat_common(fd, filename, argv, envp, flags); |
ba2d01629
|
1841 |
} |
0e028465d
|
1842 |
#endif |
ba2d01629
|
1843 |
|
964ee7df9
|
1844 |
void set_binfmt(struct linux_binfmt *new) |
1da177e4c
|
1845 |
{ |
801460d0c
|
1846 1847 1848 1849 |
struct mm_struct *mm = current->mm; if (mm->binfmt) module_put(mm->binfmt->module); |
1da177e4c
|
1850 |
|
801460d0c
|
1851 |
mm->binfmt = new; |
964ee7df9
|
1852 1853 |
if (new) __module_get(new->module); |
1da177e4c
|
1854 |
} |
1da177e4c
|
1855 |
EXPORT_SYMBOL(set_binfmt); |
6c5d52382
|
1856 |
/* |
7288e1187
|
1857 |
* set_dumpable stores three-value SUID_DUMP_* into mm->flags. |
6c5d52382
|
1858 1859 1860 |
*/ void set_dumpable(struct mm_struct *mm, int value) { |
7288e1187
|
1861 1862 |
if (WARN_ON((unsigned)value > SUID_DUMP_ROOT)) return; |
26e152252
|
1863 |
set_mask_bits(&mm->flags, MMF_DUMPABLE_MASK, value); |
6c5d52382
|
1864 |
} |
6c5d52382
|
1865 |
|
38b983b34
|
1866 1867 1868 1869 1870 |
SYSCALL_DEFINE3(execve, const char __user *, filename, const char __user *const __user *, argv, const char __user *const __user *, envp) { |
c4ad8f98b
|
1871 |
return do_execve(getname(filename), argv, envp); |
38b983b34
|
1872 |
} |
51f39a1f0
|
1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 |
SYSCALL_DEFINE5(execveat, int, fd, const char __user *, filename, const char __user *const __user *, argv, const char __user *const __user *, envp, int, flags) { int lookup_flags = (flags & AT_EMPTY_PATH) ? LOOKUP_EMPTY : 0; return do_execveat(fd, getname_flags(filename, lookup_flags, NULL), argv, envp, flags); } |
38b983b34
|
1886 |
#ifdef CONFIG_COMPAT |
625b1d7e8
|
1887 1888 1889 |
COMPAT_SYSCALL_DEFINE3(execve, const char __user *, filename, const compat_uptr_t __user *, argv, const compat_uptr_t __user *, envp) |
38b983b34
|
1890 |
{ |
c4ad8f98b
|
1891 |
return compat_do_execve(getname(filename), argv, envp); |
38b983b34
|
1892 |
} |
51f39a1f0
|
1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 |
COMPAT_SYSCALL_DEFINE5(execveat, int, fd, const char __user *, filename, const compat_uptr_t __user *, argv, const compat_uptr_t __user *, envp, int, flags) { int lookup_flags = (flags & AT_EMPTY_PATH) ? LOOKUP_EMPTY : 0; return compat_do_execveat(fd, getname_flags(filename, lookup_flags, NULL), argv, envp, flags); } |
38b983b34
|
1906 |
#endif |