Blame view
fs/exec.c
49.1 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 |
*/ |
b89999d00
|
25 |
#include <linux/kernel_read_file.h> |
1da177e4c
|
26 27 |
#include <linux/slab.h> #include <linux/file.h> |
9f3acc314
|
28 |
#include <linux/fdtable.h> |
ba92a43db
|
29 |
#include <linux/mm.h> |
615d6e875
|
30 |
#include <linux/vmacache.h> |
1da177e4c
|
31 32 |
#include <linux/stat.h> #include <linux/fcntl.h> |
ba92a43db
|
33 |
#include <linux/swap.h> |
74aadce98
|
34 |
#include <linux/string.h> |
1da177e4c
|
35 |
#include <linux/init.h> |
6e84f3152
|
36 |
#include <linux/sched/mm.h> |
f7ccbae45
|
37 |
#include <linux/sched/coredump.h> |
3f07c0144
|
38 |
#include <linux/sched/signal.h> |
6a3827d75
|
39 |
#include <linux/sched/numa_balancing.h> |
299300258
|
40 |
#include <linux/sched/task.h> |
ca5b172bd
|
41 |
#include <linux/pagemap.h> |
cdd6c482c
|
42 |
#include <linux/perf_event.h> |
1da177e4c
|
43 44 45 46 47 |
#include <linux/highmem.h> #include <linux/spinlock.h> #include <linux/key.h> #include <linux/personality.h> #include <linux/binfmts.h> |
1da177e4c
|
48 |
#include <linux/utsname.h> |
84d737866
|
49 |
#include <linux/pid_namespace.h> |
1da177e4c
|
50 51 |
#include <linux/module.h> #include <linux/namei.h> |
1da177e4c
|
52 53 54 |
#include <linux/mount.h> #include <linux/security.h> #include <linux/syscalls.h> |
8f0ab5147
|
55 |
#include <linux/tsacct_kern.h> |
9f46080c4
|
56 |
#include <linux/cn_proc.h> |
473ae30bc
|
57 |
#include <linux/audit.h> |
6341c393f
|
58 |
#include <linux/tracehook.h> |
5f4123be3
|
59 |
#include <linux/kmod.h> |
6110e3abb
|
60 |
#include <linux/fsnotify.h> |
5ad4e53bd
|
61 |
#include <linux/fs_struct.h> |
3d5992d2a
|
62 |
#include <linux/oom.h> |
0e028465d
|
63 |
#include <linux/compat.h> |
b44a7dfc6
|
64 |
#include <linux/vmalloc.h> |
0f2122045
|
65 |
#include <linux/io_uring.h> |
1da177e4c
|
66 |
|
7c0f6ba68
|
67 |
#include <linux/uaccess.h> |
1da177e4c
|
68 |
#include <asm/mmu_context.h> |
b6a2fea39
|
69 |
#include <asm/tlb.h> |
43d2b1132
|
70 71 |
#include <trace/events/task.h> |
a6f76f23d
|
72 |
#include "internal.h" |
1da177e4c
|
73 |
|
4ff16c25e
|
74 |
#include <trace/events/sched.h> |
56305aa9b
|
75 |
static int bprm_creds_from_file(struct linux_binprm *bprm); |
d6e711448
|
76 |
int suid_dumpable = 0; |
e4dc1b14d
|
77 |
static LIST_HEAD(formats); |
1da177e4c
|
78 |
static DEFINE_RWLOCK(binfmt_lock); |
8fc3dc5a3
|
79 |
void __register_binfmt(struct linux_binfmt * fmt, int insert) |
1da177e4c
|
80 |
{ |
8fc3dc5a3
|
81 |
BUG_ON(!fmt); |
92eaa565a
|
82 83 |
if (WARN_ON(!fmt->load_binary)) return; |
1da177e4c
|
84 |
write_lock(&binfmt_lock); |
74641f584
|
85 86 |
insert ? list_add(&fmt->lh, &formats) : list_add_tail(&fmt->lh, &formats); |
1da177e4c
|
87 |
write_unlock(&binfmt_lock); |
1da177e4c
|
88 |
} |
74641f584
|
89 |
EXPORT_SYMBOL(__register_binfmt); |
1da177e4c
|
90 |
|
f6b450d48
|
91 |
void unregister_binfmt(struct linux_binfmt * fmt) |
1da177e4c
|
92 |
{ |
1da177e4c
|
93 |
write_lock(&binfmt_lock); |
e4dc1b14d
|
94 |
list_del(&fmt->lh); |
1da177e4c
|
95 |
write_unlock(&binfmt_lock); |
1da177e4c
|
96 97 98 99 100 101 102 103 |
} EXPORT_SYMBOL(unregister_binfmt); static inline void put_binfmt(struct linux_binfmt * fmt) { module_put(fmt->module); } |
90f8572b0
|
104 105 106 107 108 |
bool path_noexec(const struct path *path) { return (path->mnt->mnt_flags & MNT_NOEXEC) || (path->mnt->mnt_sb->s_iflags & SB_I_NOEXEC); } |
69369a700
|
109 |
#ifdef CONFIG_USELIB |
1da177e4c
|
110 111 112 113 114 115 |
/* * 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
|
116 |
SYSCALL_DEFINE1(uselib, const char __user *, library) |
1da177e4c
|
117 |
{ |
72c2d5319
|
118 |
struct linux_binfmt *fmt; |
964bd1836
|
119 |
struct file *file; |
91a27b2a7
|
120 |
struct filename *tmp = getname(library); |
964bd1836
|
121 |
int error = PTR_ERR(tmp); |
47c805dc2
|
122 123 |
static const struct open_flags uselib_flags = { .open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, |
62fb4a155
|
124 |
.acc_mode = MAY_READ | MAY_EXEC, |
f9652e10c
|
125 126 |
.intent = LOOKUP_OPEN, .lookup_flags = LOOKUP_FOLLOW, |
47c805dc2
|
127 |
}; |
964bd1836
|
128 |
|
6e8341a11
|
129 130 |
if (IS_ERR(tmp)) goto out; |
f9652e10c
|
131 |
file = do_filp_open(AT_FDCWD, tmp, &uselib_flags); |
6e8341a11
|
132 133 134 |
putname(tmp); error = PTR_ERR(file); if (IS_ERR(file)) |
1da177e4c
|
135 |
goto out; |
633fb6ac3
|
136 137 138 139 140 |
/* * may_open() has already checked for this, so it should be * impossible to trip now. But we need to be extra cautious * and check again at the very end too. */ |
db19c91c3
|
141 |
error = -EACCES; |
0fd338b2d
|
142 143 |
if (WARN_ON_ONCE(!S_ISREG(file_inode(file)->i_mode) || path_noexec(&file->f_path))) |
6146f0d5e
|
144 |
goto exit; |
1da177e4c
|
145 |
|
2a12a9d78
|
146 |
fsnotify_open(file); |
6110e3abb
|
147 |
|
1da177e4c
|
148 |
error = -ENOEXEC; |
1da177e4c
|
149 |
|
72c2d5319
|
150 151 152 153 154 155 |
read_lock(&binfmt_lock); list_for_each_entry(fmt, &formats, lh) { if (!fmt->load_shlib) continue; if (!try_module_get(fmt->module)) continue; |
1da177e4c
|
156 |
read_unlock(&binfmt_lock); |
72c2d5319
|
157 158 159 160 161 |
error = fmt->load_shlib(file); read_lock(&binfmt_lock); put_binfmt(fmt); if (error != -ENOEXEC) break; |
1da177e4c
|
162 |
} |
72c2d5319
|
163 |
read_unlock(&binfmt_lock); |
6e8341a11
|
164 |
exit: |
1da177e4c
|
165 166 167 |
fput(file); out: return error; |
1da177e4c
|
168 |
} |
69369a700
|
169 |
#endif /* #ifdef CONFIG_USELIB */ |
1da177e4c
|
170 |
|
b6a2fea39
|
171 |
#ifdef CONFIG_MMU |
ae6b585ee
|
172 173 174 175 176 177 |
/* * 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
|
178 |
static void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) |
3c77f8457
|
179 180 181 182 183 184 185 186 |
{ struct mm_struct *mm = current->mm; long diff = (long)(pages - bprm->vma_pages); if (!mm || !diff) return; bprm->vma_pages = pages; |
3c77f8457
|
187 |
add_mm_counter(mm, MM_ANONPAGES, diff); |
3c77f8457
|
188 |
} |
0e028465d
|
189 |
static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, |
b6a2fea39
|
190 191 192 193 |
int write) { struct page *page; int ret; |
9beae1ea8
|
194 |
unsigned int gup_flags = FOLL_FORCE; |
b6a2fea39
|
195 196 197 |
#ifdef CONFIG_STACK_GROWSUP if (write) { |
d05f3169c
|
198 |
ret = expand_downwards(bprm->vma, pos); |
b6a2fea39
|
199 200 201 202 |
if (ret < 0) return NULL; } #endif |
9beae1ea8
|
203 204 205 |
if (write) gup_flags |= FOLL_WRITE; |
1e9877902
|
206 207 208 209 |
/* * We are doing an exec(). 'current' is the process * doing the exec and bprm->mm is the new process's mm. */ |
64019a2e4
|
210 |
ret = get_user_pages_remote(bprm->mm, pos, 1, gup_flags, |
5b56d49fc
|
211 |
&page, NULL, NULL); |
b6a2fea39
|
212 213 |
if (ret <= 0) return NULL; |
655c16a8c
|
214 215 |
if (write) acct_arg_size(bprm, vma_pages(bprm->vma)); |
b6a2fea39
|
216 217 218 219 220 221 222 223 |
return page; } static void put_arg_page(struct page *page) { put_page(page); } |
b6a2fea39
|
224 225 226 227 228 229 230 231 232 233 234 235 |
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
|
236 |
int err; |
b6a2fea39
|
237 238 |
struct vm_area_struct *vma = NULL; struct mm_struct *mm = bprm->mm; |
490fc0538
|
239 |
bprm->vma = vma = vm_area_alloc(mm); |
b6a2fea39
|
240 |
if (!vma) |
eaccbfa56
|
241 |
return -ENOMEM; |
bfd40eaff
|
242 |
vma_set_anonymous(vma); |
b6a2fea39
|
243 |
|
d8ed45c5d
|
244 |
if (mmap_write_lock_killable(mm)) { |
f268dfe90
|
245 246 247 |
err = -EINTR; goto err_free; } |
b6a2fea39
|
248 249 250 251 252 253 254 |
/* * 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
|
255 |
BUILD_BUG_ON(VM_STACK_FLAGS & VM_STACK_INCOMPLETE_SETUP); |
b6a2fea39
|
256 257 |
vma->vm_end = STACK_TOP_MAX; vma->vm_start = vma->vm_end - PAGE_SIZE; |
d9104d1ca
|
258 |
vma->vm_flags = VM_SOFTDIRTY | VM_STACK_FLAGS | VM_STACK_INCOMPLETE_SETUP; |
3ed75eb8f
|
259 |
vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); |
462e635e5
|
260 |
|
b6a2fea39
|
261 |
err = insert_vm_struct(mm, vma); |
eaccbfa56
|
262 |
if (err) |
b6a2fea39
|
263 |
goto err; |
b6a2fea39
|
264 265 |
mm->stack_vm = mm->total_vm = 1; |
d8ed45c5d
|
266 |
mmap_write_unlock(mm); |
b6a2fea39
|
267 |
bprm->p = vma->vm_end - sizeof(void *); |
b6a2fea39
|
268 |
return 0; |
b6a2fea39
|
269 |
err: |
d8ed45c5d
|
270 |
mmap_write_unlock(mm); |
f268dfe90
|
271 |
err_free: |
eaccbfa56
|
272 |
bprm->vma = NULL; |
3928d4f5e
|
273 |
vm_area_free(vma); |
b6a2fea39
|
274 275 276 277 278 279 280 281 282 |
return err; } static bool valid_arg_len(struct linux_binprm *bprm, long len) { return len <= MAX_ARG_STRLEN; } #else |
0e028465d
|
283 |
static inline void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) |
3c77f8457
|
284 285 |
{ } |
0e028465d
|
286 |
static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, |
b6a2fea39
|
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 |
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
|
346 |
static int bprm_mm_init(struct linux_binprm *bprm) |
b6a2fea39
|
347 348 349 350 351 352 353 354 |
{ int err; struct mm_struct *mm = NULL; bprm->mm = mm = mm_alloc(); err = -ENOMEM; if (!mm) goto err; |
c31dbb146
|
355 356 357 358 |
/* 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
|
359 360 361 362 363 364 365 366 367 368 369 370 371 372 |
err = __bprm_mm_init(bprm); if (err) goto err; return 0; err: if (mm) { bprm->mm = NULL; mmdrop(mm); } return err; } |
ba2d01629
|
373 |
struct user_arg_ptr { |
0e028465d
|
374 375 376 377 378 379 |
#ifdef CONFIG_COMPAT bool is_compat; #endif union { const char __user *const __user *native; #ifdef CONFIG_COMPAT |
38b983b34
|
380 |
const compat_uptr_t __user *compat; |
0e028465d
|
381 382 |
#endif } ptr; |
ba2d01629
|
383 384 385 |
}; static const char __user *get_user_arg_ptr(struct user_arg_ptr argv, int nr) |
1d1dbf813
|
386 |
{ |
0e028465d
|
387 388 389 390 391 392 393 394 |
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
|
395 |
|
0e028465d
|
396 397 398 399 400 |
return compat_ptr(compat); } #endif if (get_user(native, argv.ptr.native + nr)) |
1d1dbf813
|
401 |
return ERR_PTR(-EFAULT); |
0e028465d
|
402 |
return native; |
1d1dbf813
|
403 |
} |
1da177e4c
|
404 405 406 |
/* * count() counts the number of strings in array ARGV. */ |
ba2d01629
|
407 |
static int count(struct user_arg_ptr argv, int max) |
1da177e4c
|
408 409 |
{ int i = 0; |
0e028465d
|
410 |
if (argv.ptr.native != NULL) { |
1da177e4c
|
411 |
for (;;) { |
1d1dbf813
|
412 |
const char __user *p = get_user_arg_ptr(argv, i); |
1da177e4c
|
413 |
|
1da177e4c
|
414 415 |
if (!p) break; |
1d1dbf813
|
416 417 418 |
if (IS_ERR(p)) return -EFAULT; |
6d92d4f6a
|
419 |
if (i >= max) |
1da177e4c
|
420 |
return -E2BIG; |
6d92d4f6a
|
421 |
++i; |
9aea5a65a
|
422 423 424 |
if (fatal_signal_pending(current)) return -ERESTARTNOHAND; |
1da177e4c
|
425 426 427 428 429 |
cond_resched(); } } return i; } |
be619f7f0
|
430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 |
static int count_strings_kernel(const char *const *argv) { int i; if (!argv) return 0; for (i = 0; argv[i]; ++i) { if (i >= MAX_ARG_STRINGS) return -E2BIG; if (fatal_signal_pending(current)) return -ERESTARTNOHAND; cond_resched(); } return i; } |
d8b9cd549
|
446 |
static int bprm_stack_limits(struct linux_binprm *bprm) |
655c16a8c
|
447 448 |
{ unsigned long limit, ptr_size; |
655c16a8c
|
449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 |
/* * 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
|
479 |
/* |
b6a2fea39
|
480 481 482 |
* '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
|
483 |
*/ |
ba2d01629
|
484 |
static int copy_strings(int argc, struct user_arg_ptr argv, |
75c96f858
|
485 |
struct linux_binprm *bprm) |
1da177e4c
|
486 487 488 |
{ struct page *kmapped_page = NULL; char *kaddr = NULL; |
b6a2fea39
|
489 |
unsigned long kpos = 0; |
1da177e4c
|
490 491 492 |
int ret; while (argc-- > 0) { |
d7627467b
|
493 |
const char __user *str; |
1da177e4c
|
494 495 |
int len; unsigned long pos; |
1d1dbf813
|
496 497 498 |
ret = -EFAULT; str = get_user_arg_ptr(argv, argc); if (IS_ERR(str)) |
1da177e4c
|
499 |
goto out; |
1da177e4c
|
500 |
|
1d1dbf813
|
501 502 503 504 505 506 |
len = strnlen_user(str, MAX_ARG_STRLEN); if (!len) goto out; ret = -E2BIG; if (!valid_arg_len(bprm, len)) |
1da177e4c
|
507 |
goto out; |
1da177e4c
|
508 |
|
b6a2fea39
|
509 |
/* We're going to work our way backwords. */ |
1da177e4c
|
510 |
pos = bprm->p; |
b6a2fea39
|
511 512 |
str += len; bprm->p -= len; |
655c16a8c
|
513 514 515 516 |
#ifdef CONFIG_MMU if (bprm->p < bprm->argmin) goto out; #endif |
1da177e4c
|
517 518 |
while (len > 0) { |
1da177e4c
|
519 |
int offset, bytes_to_copy; |
1da177e4c
|
520 |
|
9aea5a65a
|
521 522 523 524 |
if (fatal_signal_pending(current)) { ret = -ERESTARTNOHAND; goto out; } |
7993bc1f4
|
525 |
cond_resched(); |
1da177e4c
|
526 |
offset = pos % PAGE_SIZE; |
b6a2fea39
|
527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 |
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
|
543 |
if (!page) { |
b6a2fea39
|
544 |
ret = -E2BIG; |
1da177e4c
|
545 546 |
goto out; } |
1da177e4c
|
547 |
|
b6a2fea39
|
548 549 |
if (kmapped_page) { flush_kernel_dcache_page(kmapped_page); |
1da177e4c
|
550 |
kunmap(kmapped_page); |
b6a2fea39
|
551 552 |
put_arg_page(kmapped_page); } |
1da177e4c
|
553 554 |
kmapped_page = page; kaddr = kmap(kmapped_page); |
b6a2fea39
|
555 556 |
kpos = pos & PAGE_MASK; flush_arg_page(bprm, kpos, kmapped_page); |
1da177e4c
|
557 |
} |
b6a2fea39
|
558 |
if (copy_from_user(kaddr+offset, str, bytes_to_copy)) { |
1da177e4c
|
559 560 561 |
ret = -EFAULT; goto out; } |
1da177e4c
|
562 563 564 565 |
} } ret = 0; out: |
b6a2fea39
|
566 567 |
if (kmapped_page) { flush_kernel_dcache_page(kmapped_page); |
1da177e4c
|
568 |
kunmap(kmapped_page); |
b6a2fea39
|
569 570 |
put_arg_page(kmapped_page); } |
1da177e4c
|
571 572 573 574 |
return ret; } /* |
986db2d14
|
575 |
* Copy and argument/environment string from the kernel to the processes stack. |
1da177e4c
|
576 |
*/ |
986db2d14
|
577 |
int copy_string_kernel(const char *arg, struct linux_binprm *bprm) |
1da177e4c
|
578 |
{ |
762a3af6f
|
579 580 581 582 583 584 585 586 587 588 589 590 591 |
int len = strnlen(arg, MAX_ARG_STRLEN) + 1 /* terminating NUL */; unsigned long pos = bprm->p; if (len == 0) return -EFAULT; if (!valid_arg_len(bprm, len)) return -E2BIG; /* We're going to work our way backwards. */ arg += len; bprm->p -= len; if (IS_ENABLED(CONFIG_MMU) && bprm->p < bprm->argmin) return -E2BIG; |
ba2d01629
|
592 |
|
762a3af6f
|
593 594 595 596 597 |
while (len > 0) { unsigned int bytes_to_copy = min_t(unsigned int, len, min_not_zero(offset_in_page(pos), PAGE_SIZE)); struct page *page; char *kaddr; |
ba2d01629
|
598 |
|
762a3af6f
|
599 600 601 |
pos -= bytes_to_copy; arg -= bytes_to_copy; len -= bytes_to_copy; |
ba2d01629
|
602 |
|
762a3af6f
|
603 604 605 606 607 608 609 610 611 612 613 614 |
page = get_arg_page(bprm, pos, 1); if (!page) return -E2BIG; kaddr = kmap_atomic(page); flush_arg_page(bprm, pos & PAGE_MASK, page); memcpy(kaddr + offset_in_page(pos), arg, bytes_to_copy); flush_kernel_dcache_page(page); kunmap_atomic(kaddr); put_arg_page(page); } return 0; |
1da177e4c
|
615 |
} |
986db2d14
|
616 |
EXPORT_SYMBOL(copy_string_kernel); |
1da177e4c
|
617 |
|
be619f7f0
|
618 619 620 621 622 623 624 625 626 627 628 629 630 |
static int copy_strings_kernel(int argc, const char *const *argv, struct linux_binprm *bprm) { while (argc-- > 0) { int ret = copy_string_kernel(argv[argc], bprm); if (ret < 0) return ret; if (fatal_signal_pending(current)) return -ERESTARTNOHAND; cond_resched(); } return 0; } |
1da177e4c
|
631 |
#ifdef CONFIG_MMU |
b6a2fea39
|
632 |
|
1da177e4c
|
633 |
/* |
b6a2fea39
|
634 635 636 |
* 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
|
637 |
* |
b6a2fea39
|
638 639 640 641 642 643 |
* 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
|
644 |
*/ |
b6a2fea39
|
645 |
static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) |
1da177e4c
|
646 647 |
{ struct mm_struct *mm = vma->vm_mm; |
b6a2fea39
|
648 649 650 651 652 |
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
|
653 |
struct mmu_gather tlb; |
1da177e4c
|
654 |
|
b6a2fea39
|
655 |
BUG_ON(new_start > new_end); |
1da177e4c
|
656 |
|
b6a2fea39
|
657 658 659 660 661 662 663 664 665 666 |
/* * 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
|
667 668 |
if (vma_adjust(vma, new_start, old_end, vma->vm_pgoff, NULL)) return -ENOMEM; |
b6a2fea39
|
669 670 671 672 673 674 |
/* * 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
|
675 |
vma, new_start, length, false)) |
b6a2fea39
|
676 677 678 |
return -ENOMEM; lru_add_drain(); |
2b047252d
|
679 |
tlb_gather_mmu(&tlb, mm, old_start, old_end); |
b6a2fea39
|
680 681 682 683 |
if (new_end > old_start) { /* * when the old and new regions overlap clear from new_end. */ |
d16dfc550
|
684 |
free_pgd_range(&tlb, new_end, old_end, new_end, |
6ee8630e0
|
685 |
vma->vm_next ? vma->vm_next->vm_start : USER_PGTABLES_CEILING); |
b6a2fea39
|
686 687 688 689 690 691 692 |
} 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
|
693 |
free_pgd_range(&tlb, old_start, old_end, new_end, |
6ee8630e0
|
694 |
vma->vm_next ? vma->vm_next->vm_start : USER_PGTABLES_CEILING); |
1da177e4c
|
695 |
} |
2b047252d
|
696 |
tlb_finish_mmu(&tlb, old_start, old_end); |
b6a2fea39
|
697 698 |
/* |
5beb49305
|
699 |
* Shrink the vma to just the new range. Always succeeds. |
b6a2fea39
|
700 701 702 703 |
*/ vma_adjust(vma, new_start, new_end, vma->vm_pgoff, NULL); return 0; |
1da177e4c
|
704 |
} |
b6a2fea39
|
705 706 707 708 |
/* * Finalizes the stack vm_area_struct. The flags and permissions are updated, * the stack is optionally relocated, and some extra space is added. */ |
1da177e4c
|
709 710 711 712 |
int setup_arg_pages(struct linux_binprm *bprm, unsigned long stack_top, int executable_stack) { |
b6a2fea39
|
713 714 |
unsigned long ret; unsigned long stack_shift; |
1da177e4c
|
715 |
struct mm_struct *mm = current->mm; |
b6a2fea39
|
716 717 718 719 |
struct vm_area_struct *vma = bprm->vma; struct vm_area_struct *prev = NULL; unsigned long vm_flags; unsigned long stack_base; |
803bf5ec2
|
720 721 722 |
unsigned long stack_size; unsigned long stack_expand; unsigned long rlim_stack; |
1da177e4c
|
723 724 |
#ifdef CONFIG_STACK_GROWSUP |
d71f290b4
|
725 |
/* Limit stack size */ |
c31dbb146
|
726 |
stack_base = bprm->rlim_stack.rlim_max; |
d71f290b4
|
727 728 |
if (stack_base > STACK_SIZE_MAX) stack_base = STACK_SIZE_MAX; |
1da177e4c
|
729 |
|
d045c77c1
|
730 731 |
/* Add space for stack randomization. */ stack_base += (STACK_RND_MASK << PAGE_SHIFT); |
b6a2fea39
|
732 733 734 |
/* Make sure we didn't let the argument array grow too large. */ if (vma->vm_end - vma->vm_start > stack_base) return -ENOMEM; |
1da177e4c
|
735 |
|
b6a2fea39
|
736 |
stack_base = PAGE_ALIGN(stack_top - stack_base); |
1da177e4c
|
737 |
|
b6a2fea39
|
738 739 740 |
stack_shift = vma->vm_start - stack_base; mm->arg_start = bprm->p - stack_shift; bprm->p = vma->vm_end - stack_shift; |
1da177e4c
|
741 |
#else |
b6a2fea39
|
742 743 |
stack_top = arch_align_stack(stack_top); stack_top = PAGE_ALIGN(stack_top); |
1b528181b
|
744 745 746 747 |
if (unlikely(stack_top < mmap_min_addr) || unlikely(vma->vm_end - vma->vm_start >= stack_top - mmap_min_addr)) return -ENOMEM; |
b6a2fea39
|
748 749 750 |
stack_shift = vma->vm_end - stack_top; bprm->p -= stack_shift; |
1da177e4c
|
751 |
mm->arg_start = bprm->p; |
1da177e4c
|
752 |
#endif |
1da177e4c
|
753 |
if (bprm->loader) |
b6a2fea39
|
754 755 |
bprm->loader -= stack_shift; bprm->exec -= stack_shift; |
1da177e4c
|
756 |
|
d8ed45c5d
|
757 |
if (mmap_write_lock_killable(mm)) |
f268dfe90
|
758 |
return -EINTR; |
96a8e13ed
|
759 |
vm_flags = VM_STACK_FLAGS; |
b6a2fea39
|
760 761 762 763 764 765 766 767 768 769 770 |
/* * 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
|
771 |
vm_flags |= VM_STACK_INCOMPLETE_SETUP; |
b6a2fea39
|
772 773 774 775 776 777 |
ret = mprotect_fixup(vma, &prev, vma->vm_start, vma->vm_end, vm_flags); if (ret) goto out_unlock; BUG_ON(prev != vma); |
47a2ebb7f
|
778 779 780 781 782 |
if (unlikely(vm_flags & VM_EXEC)) { pr_warn_once("process '%pD4' started with executable stack ", bprm->file); } |
b6a2fea39
|
783 784 785 |
/* Move stack pages down in memory. */ if (stack_shift) { ret = shift_arg_pages(vma, stack_shift); |
fc63cf237
|
786 787 |
if (ret) goto out_unlock; |
1da177e4c
|
788 |
} |
a8bef8ff6
|
789 790 |
/* mprotect_fixup is overkill to remove the temporary stack flags */ vma->vm_flags &= ~VM_STACK_INCOMPLETE_SETUP; |
5ef097dd7
|
791 |
stack_expand = 131072UL; /* randomly 32*4k (or 2*64k) pages */ |
803bf5ec2
|
792 793 794 795 796 |
stack_size = vma->vm_end - vma->vm_start; /* * Align this down to a page boundary as expand_stack * will align it up. */ |
c31dbb146
|
797 |
rlim_stack = bprm->rlim_stack.rlim_cur & PAGE_MASK; |
b6a2fea39
|
798 |
#ifdef CONFIG_STACK_GROWSUP |
803bf5ec2
|
799 800 801 802 |
if (stack_size + stack_expand > rlim_stack) stack_base = vma->vm_start + rlim_stack; else stack_base = vma->vm_end + stack_expand; |
b6a2fea39
|
803 |
#else |
803bf5ec2
|
804 805 806 807 |
if (stack_size + stack_expand > rlim_stack) stack_base = vma->vm_end - rlim_stack; else stack_base = vma->vm_start - stack_expand; |
b6a2fea39
|
808 |
#endif |
3af9e8592
|
809 |
current->mm->start_stack = bprm->p; |
b6a2fea39
|
810 811 812 813 814 |
ret = expand_stack(vma, stack_base); if (ret) ret = -EFAULT; out_unlock: |
d8ed45c5d
|
815 |
mmap_write_unlock(mm); |
fc63cf237
|
816 |
return ret; |
1da177e4c
|
817 |
} |
1da177e4c
|
818 |
EXPORT_SYMBOL(setup_arg_pages); |
7e7ec6a93
|
819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 |
#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
|
851 |
#endif /* CONFIG_MMU */ |
51f39a1f0
|
852 |
static struct file *do_open_execat(int fd, struct filename *name, int flags) |
1da177e4c
|
853 |
{ |
1da177e4c
|
854 |
struct file *file; |
e56b6a5dd
|
855 |
int err; |
51f39a1f0
|
856 |
struct open_flags open_exec_flags = { |
47c805dc2
|
857 |
.open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, |
62fb4a155
|
858 |
.acc_mode = MAY_EXEC, |
f9652e10c
|
859 860 |
.intent = LOOKUP_OPEN, .lookup_flags = LOOKUP_FOLLOW, |
47c805dc2
|
861 |
}; |
1da177e4c
|
862 |
|
51f39a1f0
|
863 864 865 866 867 868 869 870 |
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
|
871 |
if (IS_ERR(file)) |
e56b6a5dd
|
872 |
goto out; |
633fb6ac3
|
873 874 875 876 877 |
/* * may_open() has already checked for this, so it should be * impossible to trip now. But we need to be extra cautious * and check again at the very end too. */ |
e56b6a5dd
|
878 |
err = -EACCES; |
0fd338b2d
|
879 880 |
if (WARN_ON_ONCE(!S_ISREG(file_inode(file)->i_mode) || path_noexec(&file->f_path))) |
6e8341a11
|
881 |
goto exit; |
e56b6a5dd
|
882 883 |
err = deny_write_access(file); |
6e8341a11
|
884 885 |
if (err) goto exit; |
1da177e4c
|
886 |
|
51f39a1f0
|
887 888 |
if (name->name[0] != '\0') fsnotify_open(file); |
6e8341a11
|
889 |
out: |
e56b6a5dd
|
890 |
return file; |
6e8341a11
|
891 892 |
exit: fput(file); |
e56b6a5dd
|
893 894 |
return ERR_PTR(err); } |
c4ad8f98b
|
895 896 897 |
struct file *open_exec(const char *name) { |
516891041
|
898 899 900 901 902 903 904 905 |
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
|
906 |
} |
1da177e4c
|
907 |
EXPORT_SYMBOL(open_exec); |
48304f799
|
908 909 |
#if defined(CONFIG_HAVE_AOUT) || defined(CONFIG_BINFMT_FLAT) || \ defined(CONFIG_BINFMT_ELF_FDPIC) |
3dc20cb28
|
910 911 |
ssize_t read_code(struct file *file, unsigned long addr, loff_t pos, size_t len) { |
ec6955798
|
912 |
ssize_t res = vfs_read(file, (void __user *)addr, len, &pos); |
3dc20cb28
|
913 |
if (res > 0) |
bce2b68b8
|
914 |
flush_icache_user_range(addr, addr + len); |
3dc20cb28
|
915 916 917 |
return res; } EXPORT_SYMBOL(read_code); |
48304f799
|
918 |
#endif |
3dc20cb28
|
919 |
|
eea967325
|
920 921 |
/* * Maps the mm_struct mm into the current task struct. |
ab7709b55
|
922 923 |
* On success, this function returns with exec_update_lock * held for writing. |
eea967325
|
924 |
*/ |
1da177e4c
|
925 926 927 |
static int exec_mmap(struct mm_struct *mm) { struct task_struct *tsk; |
615d6e875
|
928 |
struct mm_struct *old_mm, *active_mm; |
eea967325
|
929 |
int ret; |
1da177e4c
|
930 931 932 933 |
/* Notify parent that we're no longer interested in the old VM */ tsk = current; old_mm = current->mm; |
4610ba7ad
|
934 |
exec_mm_release(tsk, old_mm); |
a28bf136e
|
935 936 |
if (old_mm) sync_mm_rss(old_mm); |
1da177e4c
|
937 |
|
ab7709b55
|
938 |
ret = down_write_killable(&tsk->signal->exec_update_lock); |
eea967325
|
939 940 |
if (ret) return ret; |
1da177e4c
|
941 942 943 944 |
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 |
c1e8d7c6a
|
945 |
* through with the exec. We must hold mmap_lock around |
999d9fc16
|
946 |
* checking core_state and changing tsk->mm. |
1da177e4c
|
947 |
*/ |
d8ed45c5d
|
948 |
mmap_read_lock(old_mm); |
999d9fc16
|
949 |
if (unlikely(old_mm->core_state)) { |
d8ed45c5d
|
950 |
mmap_read_unlock(old_mm); |
ab7709b55
|
951 |
up_write(&tsk->signal->exec_update_lock); |
1da177e4c
|
952 953 954 |
return -EINTR; } } |
eea967325
|
955 |
|
1da177e4c
|
956 |
task_lock(tsk); |
227a4aadc
|
957 |
membarrier_exec_mmap(mm); |
d53c3dfb2
|
958 959 960 |
local_irq_disable(); active_mm = tsk->active_mm; |
1da177e4c
|
961 |
tsk->active_mm = mm; |
d53c3dfb2
|
962 963 964 965 966 967 968 969 970 971 |
tsk->mm = mm; /* * This prevents preemption while active_mm is being loaded and * it and mm are being updated, which could cause problems for * lazy tlb mm refcounting when these are updated by context * switches. Not all architectures can handle irqs off over * activate_mm yet. */ if (!IS_ENABLED(CONFIG_ARCH_WANT_IRQS_OFF_ACTIVATE_MM)) local_irq_enable(); |
1da177e4c
|
972 |
activate_mm(active_mm, mm); |
d53c3dfb2
|
973 974 |
if (IS_ENABLED(CONFIG_ARCH_WANT_IRQS_OFF_ACTIVATE_MM)) local_irq_enable(); |
615d6e875
|
975 976 |
tsk->mm->vmacache_seqnum = 0; vmacache_flush(tsk); |
1da177e4c
|
977 |
task_unlock(tsk); |
1da177e4c
|
978 |
if (old_mm) { |
d8ed45c5d
|
979 |
mmap_read_unlock(old_mm); |
7dddb12c6
|
980 |
BUG_ON(active_mm != old_mm); |
701085b21
|
981 |
setmax_mm_hiwater_rss(&tsk->signal->maxrss, old_mm); |
31a78f23b
|
982 |
mm_update_next_owner(old_mm); |
1da177e4c
|
983 984 985 986 987 988 |
mmput(old_mm); return 0; } mmdrop(active_mm); return 0; } |
858119e15
|
989 |
static int de_thread(struct task_struct *tsk) |
1da177e4c
|
990 991 |
{ struct signal_struct *sig = tsk->signal; |
b2c903b87
|
992 |
struct sighand_struct *oldsighand = tsk->sighand; |
1da177e4c
|
993 |
spinlock_t *lock = &oldsighand->siglock; |
1da177e4c
|
994 |
|
aafe6c2a2
|
995 |
if (thread_group_empty(tsk)) |
1da177e4c
|
996 997 998 999 |
goto no_thread_group; /* * Kill all other threads in the thread group. |
1da177e4c
|
1000 |
*/ |
1da177e4c
|
1001 |
spin_lock_irq(lock); |
ed5d2cac1
|
1002 |
if (signal_group_exit(sig)) { |
1da177e4c
|
1003 1004 1005 1006 1007 |
/* * Another group action in progress, just * return so that the signal is processed. */ spin_unlock_irq(lock); |
1da177e4c
|
1008 1009 |
return -EAGAIN; } |
d344193a0
|
1010 |
|
ed5d2cac1
|
1011 |
sig->group_exit_task = tsk; |
d344193a0
|
1012 1013 1014 |
sig->notify_count = zap_other_threads(tsk); if (!thread_group_leader(tsk)) sig->notify_count--; |
1da177e4c
|
1015 |
|
d344193a0
|
1016 |
while (sig->notify_count) { |
d5bbd43d5
|
1017 |
__set_current_state(TASK_KILLABLE); |
1da177e4c
|
1018 |
spin_unlock_irq(lock); |
a72173ecf
|
1019 |
schedule(); |
08d405c8b
|
1020 |
if (__fatal_signal_pending(tsk)) |
d5bbd43d5
|
1021 |
goto killed; |
1da177e4c
|
1022 1023 |
spin_lock_irq(lock); } |
1da177e4c
|
1024 1025 1026 1027 1028 1029 1030 |
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
|
1031 |
if (!thread_group_leader(tsk)) { |
8187926bd
|
1032 |
struct task_struct *leader = tsk->group_leader; |
6db840fa7
|
1033 |
|
6db840fa7
|
1034 |
for (;;) { |
780de9dd2
|
1035 |
cgroup_threadgroup_change_begin(tsk); |
6db840fa7
|
1036 |
write_lock_irq(&tasklist_lock); |
dfcce791f
|
1037 1038 1039 1040 1041 |
/* * Do this under tasklist_lock to ensure that * exit_notify() can't miss ->group_exit_task */ sig->notify_count = -1; |
6db840fa7
|
1042 1043 |
if (likely(leader->exit_state)) break; |
d5bbd43d5
|
1044 |
__set_current_state(TASK_KILLABLE); |
6db840fa7
|
1045 |
write_unlock_irq(&tasklist_lock); |
780de9dd2
|
1046 |
cgroup_threadgroup_change_end(tsk); |
a72173ecf
|
1047 |
schedule(); |
08d405c8b
|
1048 |
if (__fatal_signal_pending(tsk)) |
d5bbd43d5
|
1049 |
goto killed; |
6db840fa7
|
1050 |
} |
1da177e4c
|
1051 |
|
f5e902817
|
1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 |
/* * 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
|
1062 |
tsk->start_time = leader->start_time; |
cf25e24db
|
1063 |
tsk->start_boottime = leader->start_boottime; |
f5e902817
|
1064 |
|
bac0abd61
|
1065 |
BUG_ON(!same_thread_group(leader, tsk)); |
1da177e4c
|
1066 1067 1068 1069 1070 1071 |
/* * 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
|
1072 1073 |
/* Become a process group leader with the old leader's pid. |
c18258c6f
|
1074 |
* The old leader becomes a thread of the this thread group. |
d73d65293
|
1075 |
*/ |
6b03d1304
|
1076 |
exchange_tids(tsk, leader); |
6883f81aa
|
1077 |
transfer_pid(leader, tsk, PIDTYPE_TGID); |
aafe6c2a2
|
1078 1079 |
transfer_pid(leader, tsk, PIDTYPE_PGID); transfer_pid(leader, tsk, PIDTYPE_SID); |
9cd80bbb0
|
1080 |
|
aafe6c2a2
|
1081 |
list_replace_rcu(&leader->tasks, &tsk->tasks); |
9cd80bbb0
|
1082 |
list_replace_init(&leader->sibling, &tsk->sibling); |
1da177e4c
|
1083 |
|
aafe6c2a2
|
1084 1085 |
tsk->group_leader = tsk; leader->group_leader = tsk; |
de12a7878
|
1086 |
|
aafe6c2a2
|
1087 |
tsk->exit_signal = SIGCHLD; |
087806b12
|
1088 |
leader->exit_signal = -1; |
962b564cf
|
1089 1090 1091 |
BUG_ON(leader->exit_state != EXIT_ZOMBIE); leader->exit_state = EXIT_DEAD; |
eac1b5e57
|
1092 1093 1094 1095 1096 1097 1098 1099 |
/* * 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
|
1100 |
write_unlock_irq(&tasklist_lock); |
780de9dd2
|
1101 |
cgroup_threadgroup_change_end(tsk); |
8187926bd
|
1102 1103 |
release_task(leader); |
ed5d2cac1
|
1104 |
} |
1da177e4c
|
1105 |
|
6db840fa7
|
1106 1107 |
sig->group_exit_task = NULL; sig->notify_count = 0; |
1da177e4c
|
1108 1109 |
no_thread_group: |
e63682534
|
1110 1111 |
/* we have changed execution domain */ tsk->exit_signal = SIGCHLD; |
021691559
|
1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 |
BUG_ON(!thread_group_leader(tsk)); return 0; 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; } |
7a60ef480
|
1123 1124 1125 1126 1127 1128 |
/* * 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().) */ |
021691559
|
1129 1130 1131 |
static int unshare_sighand(struct task_struct *me) { struct sighand_struct *oldsighand = me->sighand; |
329f7dba5
|
1132 |
|
d036bda7d
|
1133 |
if (refcount_read(&oldsighand->count) != 1) { |
b2c903b87
|
1134 |
struct sighand_struct *newsighand; |
1da177e4c
|
1135 |
/* |
b2c903b87
|
1136 1137 |
* This ->sighand is shared with the CLONE_SIGHAND * but not CLONE_THREAD task, switch to the new one. |
1da177e4c
|
1138 |
*/ |
b2c903b87
|
1139 1140 1141 |
newsighand = kmem_cache_alloc(sighand_cachep, GFP_KERNEL); if (!newsighand) return -ENOMEM; |
d036bda7d
|
1142 |
refcount_set(&newsighand->count, 1); |
1da177e4c
|
1143 1144 1145 1146 1147 |
memcpy(newsighand->action, oldsighand->action, sizeof(newsighand->action)); write_lock_irq(&tasklist_lock); spin_lock(&oldsighand->siglock); |
021691559
|
1148 |
rcu_assign_pointer(me->sighand, newsighand); |
1da177e4c
|
1149 1150 |
spin_unlock(&oldsighand->siglock); write_unlock_irq(&tasklist_lock); |
fba2afaae
|
1151 |
__cleanup_sighand(oldsighand); |
1da177e4c
|
1152 |
} |
1da177e4c
|
1153 1154 |
return 0; } |
0840a90d9
|
1155 |
|
3756f6401
|
1156 |
char *__get_task_comm(char *buf, size_t buf_size, struct task_struct *tsk) |
1da177e4c
|
1157 |
{ |
1da177e4c
|
1158 |
task_lock(tsk); |
3756f6401
|
1159 |
strncpy(buf, tsk->comm, buf_size); |
1da177e4c
|
1160 |
task_unlock(tsk); |
59714d65d
|
1161 |
return buf; |
1da177e4c
|
1162 |
} |
3756f6401
|
1163 |
EXPORT_SYMBOL_GPL(__get_task_comm); |
1da177e4c
|
1164 |
|
6a6d27de3
|
1165 1166 1167 1168 |
/* * These functions flushes out all traces of the currently running executable * so that a new one can be started */ |
82b897782
|
1169 |
void __set_task_comm(struct task_struct *tsk, const char *buf, bool exec) |
1da177e4c
|
1170 1171 |
{ task_lock(tsk); |
43d2b1132
|
1172 |
trace_task_rename(tsk, buf); |
1da177e4c
|
1173 1174 |
strlcpy(tsk->comm, buf, sizeof(tsk->comm)); task_unlock(tsk); |
82b897782
|
1175 |
perf_event_comm(tsk, exec); |
1da177e4c
|
1176 |
} |
a9208e42b
|
1177 1178 1179 1180 |
/* * 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 |
13c432b51
|
1181 |
* (after exec_mmap()) by search_binary_handler (see below). |
a9208e42b
|
1182 |
*/ |
2388777a0
|
1183 |
int begin_new_exec(struct linux_binprm * bprm) |
1da177e4c
|
1184 |
{ |
2ca7be7d5
|
1185 |
struct task_struct *me = current; |
221af7f87
|
1186 |
int retval; |
1da177e4c
|
1187 |
|
56305aa9b
|
1188 1189 1190 1191 |
/* Once we are committed compute the creds */ retval = bprm_creds_from_file(bprm); if (retval) return retval; |
1da177e4c
|
1192 |
/* |
6834e0bb4
|
1193 1194 1195 |
* Ensure all future errors are fatal. */ bprm->point_of_no_return = true; |
1da177e4c
|
1196 |
/* |
021691559
|
1197 |
* Make this the only thread in the thread group. |
1da177e4c
|
1198 |
*/ |
2ca7be7d5
|
1199 |
retval = de_thread(me); |
1da177e4c
|
1200 1201 |
if (retval) goto out; |
6e399cd14
|
1202 1203 1204 1205 1206 |
/* * Must be called _before_ exec_mmap() as bprm->mm is * not visibile until then. This also enables the update * to be lockless. */ |
925d1c401
|
1207 |
set_mm_exe_file(bprm->mm, bprm->file); |
6e399cd14
|
1208 |
|
b8a61c9e7
|
1209 |
/* If the binary is not readable then enforce mm->dumpable=0 */ |
f87d1c955
|
1210 |
would_dump(bprm, bprm->file); |
b8a61c9e7
|
1211 1212 |
if (bprm->have_execfd) would_dump(bprm, bprm->executable); |
f87d1c955
|
1213 |
|
1da177e4c
|
1214 |
/* |
1da177e4c
|
1215 1216 |
* Release all of the old mmap stuff */ |
3c77f8457
|
1217 |
acct_arg_size(bprm, 0); |
1da177e4c
|
1218 1219 |
retval = exec_mmap(bprm->mm); if (retval) |
fd8328be8
|
1220 |
goto out; |
1da177e4c
|
1221 |
|
a9208e42b
|
1222 |
bprm->mm = NULL; |
7ab02af42
|
1223 |
|
ccf0fa6be
|
1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 |
#ifdef CONFIG_POSIX_TIMERS exit_itimers(me->signal); flush_itimer_signals(); #endif /* * Make the signal table private. */ retval = unshare_sighand(me); if (retval) |
89826cce3
|
1234 |
goto out_unlock; |
ccf0fa6be
|
1235 |
|
fe8141759
|
1236 1237 1238 1239 1240 |
/* * Ensure that the uaccess routines can actually operate on userspace * pointers: */ force_uaccess_begin(); |
2ca7be7d5
|
1241 |
me->flags &= ~(PF_RANDOMIZE | PF_FORKNOEXEC | PF_KTHREAD | |
b88fae644
|
1242 |
PF_NOFREEZE | PF_NO_SETAFFINITY); |
7ab02af42
|
1243 |
flush_thread(); |
2ca7be7d5
|
1244 |
me->personality &= ~bprm->per_clear; |
7ab02af42
|
1245 |
|
613cc2b6f
|
1246 1247 1248 1249 1250 1251 |
/* * 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). */ |
2ca7be7d5
|
1252 |
do_close_on_exec(me->files); |
f84df2a6f
|
1253 |
|
64701dee4
|
1254 |
if (bprm->secureexec) { |
fe8993b3a
|
1255 |
/* Make sure parent cannot signal privileged process. */ |
7d503feba
|
1256 |
me->pdeath_signal = 0; |
fe8993b3a
|
1257 |
|
64701dee4
|
1258 1259 1260 1261 1262 |
/* * 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
|
1263 |
* needing to clean up the change on failure. |
64701dee4
|
1264 |
*/ |
c31dbb146
|
1265 1266 |
if (bprm->rlim_stack.rlim_cur > _STK_LIM) bprm->rlim_stack.rlim_cur = _STK_LIM; |
64701dee4
|
1267 |
} |
7d503feba
|
1268 |
me->sas_ss_sp = me->sas_ss_size = 0; |
1da177e4c
|
1269 |
|
e816c201a
|
1270 1271 1272 1273 1274 |
/* * 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
|
1275 |
if (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP || |
e816c201a
|
1276 1277 |
!(uid_eq(current_euid(), current_uid()) && gid_eq(current_egid(), current_gid()))) |
6c5d52382
|
1278 |
set_dumpable(current->mm, suid_dumpable); |
473d89639
|
1279 1280 |
else set_dumpable(current->mm, SUID_DUMP_USER); |
d6e711448
|
1281 |
|
e041e328c
|
1282 |
perf_event_exec(); |
7d503feba
|
1283 |
__set_task_comm(me, kbasename(bprm->filename), true); |
1da177e4c
|
1284 |
|
1da177e4c
|
1285 1286 |
/* An exec changes our domain. We are no longer part of the thread group */ |
7d503feba
|
1287 1288 |
WRITE_ONCE(me->self_exec_id, me->self_exec_id + 1); flush_signal_handlers(me, 0); |
96ecee29b
|
1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 |
/* * install the new credentials for this executable */ security_bprm_committing_creds(bprm); commit_creds(bprm->cred); bprm->cred = NULL; /* * Disable monitoring for regular users * when executing setuid binaries. Must * wait until new credentials are committed * by commit_creds() above */ |
7d503feba
|
1304 1305 |
if (get_dumpable(me->mm) != SUID_DUMP_USER) perf_event_exit_task(me); |
96ecee29b
|
1306 1307 1308 1309 1310 1311 |
/* * cred_guard_mutex must be held at least to this point to prevent * ptrace_attach() from altering our determination of the task's * credentials; any time after this it may be unlocked. */ security_bprm_committed_creds(bprm); |
b8a61c9e7
|
1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 |
/* Pass the opened binary to the interpreter. */ if (bprm->have_execfd) { retval = get_unused_fd_flags(0); if (retval < 0) goto out_unlock; fd_install(retval, bprm->executable); bprm->executable = NULL; bprm->execfd = retval; } |
221af7f87
|
1322 |
return 0; |
df9e4d2c4
|
1323 |
out_unlock: |
ab7709b55
|
1324 |
up_write(&me->signal->exec_update_lock); |
221af7f87
|
1325 1326 1327 |
out: return retval; } |
2388777a0
|
1328 |
EXPORT_SYMBOL(begin_new_exec); |
221af7f87
|
1329 |
|
1b5d783c9
|
1330 1331 |
void would_dump(struct linux_binprm *bprm, struct file *file) { |
f84df2a6f
|
1332 1333 1334 |
struct inode *inode = file_inode(file); if (inode_permission(inode, MAY_READ) < 0) { struct user_namespace *old, *user_ns; |
1b5d783c9
|
1335 |
bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP; |
f84df2a6f
|
1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 |
/* 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
|
1348 1349 |
} EXPORT_SYMBOL(would_dump); |
221af7f87
|
1350 1351 |
void setup_new_exec(struct linux_binprm * bprm) { |
df9e4d2c4
|
1352 1353 |
/* Setup things that can depend upon the personality */ struct task_struct *me = current; |
1da177e4c
|
1354 |
|
df9e4d2c4
|
1355 |
arch_pick_mmap_layout(me->mm, &bprm->rlim_stack); |
d6e711448
|
1356 |
|
e9ea1e7f5
|
1357 |
arch_setup_new_exec(); |
1da177e4c
|
1358 |
|
0551fbd29
|
1359 1360 1361 1362 |
/* 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 */ |
df9e4d2c4
|
1363 |
me->mm->task_size = TASK_SIZE; |
ab7709b55
|
1364 |
up_write(&me->signal->exec_update_lock); |
7d503feba
|
1365 |
mutex_unlock(&me->signal->cred_guard_mutex); |
1da177e4c
|
1366 |
} |
221af7f87
|
1367 |
EXPORT_SYMBOL(setup_new_exec); |
1da177e4c
|
1368 |
|
b83838313
|
1369 1370 1371 |
/* Runs immediately before start_thread() takes over. */ void finalize_exec(struct linux_binprm *bprm) { |
c31dbb146
|
1372 1373 1374 1375 |
/* 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
|
1376 1377 |
} EXPORT_SYMBOL(finalize_exec); |
a6f76f23d
|
1378 |
/* |
a2a8474c3
|
1379 |
* Prepare credentials and lock ->cred_guard_mutex. |
96ecee29b
|
1380 |
* setup_new_exec() commits the new creds and drops the lock. |
a2a8474c3
|
1381 1382 1383 |
* Or, if exec fails before, free_bprm() should release ->cred and * and unlock. */ |
4addd2640
|
1384 |
static int prepare_bprm_creds(struct linux_binprm *bprm) |
a2a8474c3
|
1385 |
{ |
9b1bf12d5
|
1386 |
if (mutex_lock_interruptible(¤t->signal->cred_guard_mutex)) |
a2a8474c3
|
1387 1388 1389 1390 1391 |
return -ERESTARTNOINTR; bprm->cred = prepare_exec_creds(); if (likely(bprm->cred)) return 0; |
9b1bf12d5
|
1392 |
mutex_unlock(¤t->signal->cred_guard_mutex); |
a2a8474c3
|
1393 1394 |
return -ENOMEM; } |
c4ad8f98b
|
1395 |
static void free_bprm(struct linux_binprm *bprm) |
a2a8474c3
|
1396 |
{ |
f18ac551e
|
1397 1398 1399 1400 |
if (bprm->mm) { acct_arg_size(bprm, 0); mmput(bprm->mm); } |
a2a8474c3
|
1401 1402 |
free_arg_pages(bprm); if (bprm->cred) { |
9b1bf12d5
|
1403 |
mutex_unlock(¤t->signal->cred_guard_mutex); |
a2a8474c3
|
1404 1405 |
abort_creds(bprm->cred); } |
63e46b95e
|
1406 1407 1408 1409 |
if (bprm->file) { allow_write_access(bprm->file); fput(bprm->file); } |
b8a61c9e7
|
1410 1411 |
if (bprm->executable) fput(bprm->executable); |
b66c59840
|
1412 1413 1414 |
/* If a binfmt changed the interp, free it. */ if (bprm->interp != bprm->filename) kfree(bprm->interp); |
60d9ad1d1
|
1415 |
kfree(bprm->fdpath); |
a2a8474c3
|
1416 1417 |
kfree(bprm); } |
60d9ad1d1
|
1418 |
static struct linux_binprm *alloc_bprm(int fd, struct filename *filename) |
0a8f36eb4
|
1419 1420 |
{ struct linux_binprm *bprm = kzalloc(sizeof(*bprm), GFP_KERNEL); |
60d9ad1d1
|
1421 |
int retval = -ENOMEM; |
0a8f36eb4
|
1422 |
if (!bprm) |
60d9ad1d1
|
1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 |
goto out; if (fd == AT_FDCWD || filename->name[0] == '/') { bprm->filename = filename->name; } else { if (filename->name[0] == '\0') bprm->fdpath = kasprintf(GFP_KERNEL, "/dev/fd/%d", fd); else bprm->fdpath = kasprintf(GFP_KERNEL, "/dev/fd/%d/%s", fd, filename->name); if (!bprm->fdpath) goto out_free; bprm->filename = bprm->fdpath; } bprm->interp = bprm->filename; |
f18ac551e
|
1439 1440 1441 1442 |
retval = bprm_mm_init(bprm); if (retval) goto out_free; |
0a8f36eb4
|
1443 |
return bprm; |
60d9ad1d1
|
1444 1445 1446 1447 1448 |
out_free: free_bprm(bprm); out: return ERR_PTR(retval); |
0a8f36eb4
|
1449 |
} |
c2315c187
|
1450 |
int bprm_change_interp(const char *interp, struct linux_binprm *bprm) |
b66c59840
|
1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 |
{ /* 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
|
1461 |
/* |
a6f76f23d
|
1462 |
* determine how safe it is to execute the proposed program |
9b1bf12d5
|
1463 |
* - the caller must hold ->cred_guard_mutex to protect against |
c2e1f2e30
|
1464 |
* PTRACE_ATTACH or seccomp thread-sync |
a6f76f23d
|
1465 |
*/ |
9e00cdb09
|
1466 |
static void check_unsafe_exec(struct linux_binprm *bprm) |
a6f76f23d
|
1467 |
{ |
0bf2f3aec
|
1468 |
struct task_struct *p = current, *t; |
f1191b50e
|
1469 |
unsigned n_fs; |
a6f76f23d
|
1470 |
|
9227dd2a8
|
1471 1472 |
if (p->ptrace) bprm->unsafe |= LSM_UNSAFE_PTRACE; |
a6f76f23d
|
1473 |
|
259e5e6c7
|
1474 1475 1476 1477 |
/* * This isn't strictly necessary, but it makes it harder for LSMs to * mess up. */ |
1d4457f99
|
1478 |
if (task_no_new_privs(current)) |
259e5e6c7
|
1479 |
bprm->unsafe |= LSM_UNSAFE_NO_NEW_PRIVS; |
83f62a2ea
|
1480 |
t = p; |
0bf2f3aec
|
1481 |
n_fs = 1; |
2a4419b5b
|
1482 |
spin_lock(&p->fs->lock); |
437f7fdb6
|
1483 |
rcu_read_lock(); |
83f62a2ea
|
1484 |
while_each_thread(p, t) { |
0bf2f3aec
|
1485 1486 |
if (t->fs == p->fs) n_fs++; |
0bf2f3aec
|
1487 |
} |
437f7fdb6
|
1488 |
rcu_read_unlock(); |
0bf2f3aec
|
1489 |
|
9e00cdb09
|
1490 |
if (p->fs->users > n_fs) |
a6f76f23d
|
1491 |
bprm->unsafe |= LSM_UNSAFE_SHARE; |
9e00cdb09
|
1492 1493 |
else p->fs->in_exec = 1; |
2a4419b5b
|
1494 |
spin_unlock(&p->fs->lock); |
a6f76f23d
|
1495 |
} |
56305aa9b
|
1496 |
static void bprm_fill_uid(struct linux_binprm *bprm, struct file *file) |
8b01fc86b
|
1497 |
{ |
56305aa9b
|
1498 |
/* Handle suid and sgid on files */ |
8b01fc86b
|
1499 1500 1501 1502 |
struct inode *inode; unsigned int mode; kuid_t uid; kgid_t gid; |
56305aa9b
|
1503 |
if (!mnt_may_suid(file->f_path.mnt)) |
8b01fc86b
|
1504 1505 1506 1507 |
return; if (task_no_new_privs(current)) return; |
56305aa9b
|
1508 |
inode = file->f_path.dentry->d_inode; |
8b01fc86b
|
1509 1510 1511 1512 1513 |
mode = READ_ONCE(inode->i_mode); if (!(mode & (S_ISUID|S_ISGID))) return; /* Be careful if suid/sgid is set */ |
5955102c9
|
1514 |
inode_lock(inode); |
8b01fc86b
|
1515 1516 1517 1518 1519 |
/* reload atomically mode/uid/gid now that lock held */ mode = inode->i_mode; uid = inode->i_uid; gid = inode->i_gid; |
5955102c9
|
1520 |
inode_unlock(inode); |
8b01fc86b
|
1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 |
/* 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
|
1537 |
/* |
56305aa9b
|
1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 |
* Compute brpm->cred based upon the final binary. */ static int bprm_creds_from_file(struct linux_binprm *bprm) { /* Compute creds based on which file? */ struct file *file = bprm->execfd_creds ? bprm->executable : bprm->file; bprm_fill_uid(bprm, file); return security_bprm_creds_from_file(bprm, file); } /* |
9e00cdb09
|
1550 |
* Fill the binprm structure from the inode. |
56305aa9b
|
1551 |
* Read the first BINPRM_BUF_SIZE bytes |
a6f76f23d
|
1552 1553 |
* * This may be called multiple times for binary chains (scripts for example). |
1da177e4c
|
1554 |
*/ |
8b72ca900
|
1555 |
static int prepare_binprm(struct linux_binprm *bprm) |
1da177e4c
|
1556 |
{ |
bdd1d2d3d
|
1557 |
loff_t pos = 0; |
1da177e4c
|
1558 |
|
a6f76f23d
|
1559 |
memset(bprm->buf, 0, BINPRM_BUF_SIZE); |
bdd1d2d3d
|
1560 |
return kernel_read(bprm->file, bprm->buf, BINPRM_BUF_SIZE, &pos); |
1da177e4c
|
1561 |
} |
4fc75ff48
|
1562 1563 1564 1565 1566 |
/* * 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
|
1567 |
int remove_arg_zero(struct linux_binprm *bprm) |
1da177e4c
|
1568 |
{ |
b6a2fea39
|
1569 1570 1571 1572 |
int ret = 0; unsigned long offset; char *kaddr; struct page *page; |
4fc75ff48
|
1573 |
|
b6a2fea39
|
1574 1575 |
if (!bprm->argc) return 0; |
1da177e4c
|
1576 |
|
b6a2fea39
|
1577 1578 1579 1580 1581 1582 1583 |
do { offset = bprm->p & ~PAGE_MASK; page = get_arg_page(bprm, bprm->p, 0); if (!page) { ret = -EFAULT; goto out; } |
e8e3c3d66
|
1584 |
kaddr = kmap_atomic(page); |
4fc75ff48
|
1585 |
|
b6a2fea39
|
1586 1587 1588 |
for (; offset < PAGE_SIZE && kaddr[offset]; offset++, bprm->p++) ; |
4fc75ff48
|
1589 |
|
e8e3c3d66
|
1590 |
kunmap_atomic(kaddr); |
b6a2fea39
|
1591 |
put_arg_page(page); |
b6a2fea39
|
1592 |
} while (offset == PAGE_SIZE); |
4fc75ff48
|
1593 |
|
b6a2fea39
|
1594 1595 1596 |
bprm->p++; bprm->argc--; ret = 0; |
4fc75ff48
|
1597 |
|
b6a2fea39
|
1598 1599 |
out: return ret; |
1da177e4c
|
1600 |
} |
1da177e4c
|
1601 |
EXPORT_SYMBOL(remove_arg_zero); |
cb7b6b1cb
|
1602 1603 |
#define printable(c) (((c)=='\t') || ((c)==' ') || (0x20<=(c) && (c)<=0x7e)) |
1da177e4c
|
1604 1605 1606 |
/* * cycle the list of binary formats handler, until one recognizes the image */ |
bc2bf338d
|
1607 |
static int search_binary_handler(struct linux_binprm *bprm) |
1da177e4c
|
1608 |
{ |
cb7b6b1cb
|
1609 |
bool need_retry = IS_ENABLED(CONFIG_MODULES); |
1da177e4c
|
1610 |
struct linux_binfmt *fmt; |
cb7b6b1cb
|
1611 |
int retval; |
1da177e4c
|
1612 |
|
8b72ca900
|
1613 1614 1615 |
retval = prepare_binprm(bprm); if (retval < 0) return retval; |
d74026986
|
1616 |
|
1da177e4c
|
1617 1618 1619 |
retval = security_bprm_check(bprm); if (retval) return retval; |
1da177e4c
|
1620 |
retval = -ENOENT; |
cb7b6b1cb
|
1621 1622 1623 1624 1625 1626 |
retry: read_lock(&binfmt_lock); list_for_each_entry(fmt, &formats, lh) { if (!try_module_get(fmt->module)) continue; read_unlock(&binfmt_lock); |
d53ddd018
|
1627 |
|
cb7b6b1cb
|
1628 |
retval = fmt->load_binary(bprm); |
d53ddd018
|
1629 |
|
19d860a14
|
1630 1631 |
read_lock(&binfmt_lock); put_binfmt(fmt); |
bc2bf338d
|
1632 |
if (bprm->point_of_no_return || (retval != -ENOEXEC)) { |
19d860a14
|
1633 |
read_unlock(&binfmt_lock); |
cb7b6b1cb
|
1634 |
return retval; |
1da177e4c
|
1635 |
} |
1da177e4c
|
1636 |
} |
cb7b6b1cb
|
1637 |
read_unlock(&binfmt_lock); |
19d860a14
|
1638 |
if (need_retry) { |
cb7b6b1cb
|
1639 1640 1641 |
if (printable(bprm->buf[0]) && printable(bprm->buf[1]) && printable(bprm->buf[2]) && printable(bprm->buf[3])) return retval; |
4e0621a07
|
1642 1643 |
if (request_module("binfmt-%04x", *(ushort *)(bprm->buf + 2)) < 0) return retval; |
cb7b6b1cb
|
1644 1645 1646 |
need_retry = false; goto retry; } |
1da177e4c
|
1647 1648 |
return retval; } |
1da177e4c
|
1649 |
|
5d1baf3b6
|
1650 1651 1652 |
static int exec_binprm(struct linux_binprm *bprm) { pid_t old_pid, old_vpid; |
bc2bf338d
|
1653 |
int ret, depth; |
5d1baf3b6
|
1654 1655 1656 1657 1658 1659 |
/* 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(); |
bc2bf338d
|
1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 |
/* This allows 4 levels of binfmt rewrites before failing hard. */ for (depth = 0;; depth++) { struct file *exec; if (depth > 5) return -ELOOP; ret = search_binary_handler(bprm); if (ret < 0) return ret; if (!bprm->interpreter) break; exec = bprm->file; bprm->file = bprm->interpreter; bprm->interpreter = NULL; allow_write_access(exec); if (unlikely(bprm->have_execfd)) { if (bprm->executable) { fput(exec); return -ENOEXEC; } bprm->executable = exec; } else fput(exec); |
5d1baf3b6
|
1685 |
} |
bc2bf338d
|
1686 1687 1688 1689 1690 |
audit_bprm(bprm); trace_sched_process_exec(current, old_pid, bprm); ptrace_event(PTRACE_EVENT_EXEC, old_vpid); proc_exec_connector(current); return 0; |
5d1baf3b6
|
1691 |
} |
1da177e4c
|
1692 1693 1694 |
/* * sys_execve() executes a new program. */ |
0c9cdff05
|
1695 1696 |
static int bprm_execve(struct linux_binprm *bprm, int fd, struct filename *filename, int flags) |
1da177e4c
|
1697 |
{ |
25cf336de
|
1698 |
struct file *file; |
3b1253880
|
1699 |
struct files_struct *displaced; |
1da177e4c
|
1700 |
int retval; |
0f2122045
|
1701 1702 1703 1704 1705 |
/* * Cancel any io_uring activity across execve */ io_uring_task_cancel(); |
72fa59970
|
1706 |
|
0a8f36eb4
|
1707 1708 |
retval = unshare_files(&displaced); if (retval) |
0c9cdff05
|
1709 |
return retval; |
1da177e4c
|
1710 |
|
a2a8474c3
|
1711 1712 |
retval = prepare_bprm_creds(bprm); if (retval) |
0a8f36eb4
|
1713 |
goto out_files; |
498052bba
|
1714 |
|
9e00cdb09
|
1715 |
check_unsafe_exec(bprm); |
a2a8474c3
|
1716 |
current->in_execve = 1; |
a6f76f23d
|
1717 |
|
25cf336de
|
1718 |
file = do_open_execat(fd, filename, flags); |
1da177e4c
|
1719 1720 |
retval = PTR_ERR(file); if (IS_ERR(file)) |
498052bba
|
1721 |
goto out_unmark; |
1da177e4c
|
1722 1723 |
sched_exec(); |
1da177e4c
|
1724 |
bprm->file = file; |
60d9ad1d1
|
1725 1726 1727 1728 1729 1730 1731 1732 |
/* * 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 (bprm->fdpath && close_on_exec(fd, rcu_dereference_raw(current->files->fdt))) bprm->interp_flags |= BINPRM_FLAGS_PATH_INACCESSIBLE; |
1da177e4c
|
1733 |
|
b8bff5992
|
1734 1735 1736 |
/* Set the unchanging part of bprm->cred */ retval = security_bprm_creds_for_exec(bprm); if (retval) |
1da177e4c
|
1737 |
goto out; |
5d1baf3b6
|
1738 |
retval = exec_binprm(bprm); |
a6f76f23d
|
1739 1740 |
if (retval < 0) goto out; |
1da177e4c
|
1741 |
|
a6f76f23d
|
1742 |
/* execve succeeded */ |
498052bba
|
1743 |
current->fs->in_exec = 0; |
f9ce1f1cd
|
1744 |
current->in_execve = 0; |
d7822b1e2
|
1745 |
rseq_execve(current); |
a6f76f23d
|
1746 |
acct_update_integrals(current); |
16d51a590
|
1747 |
task_numa_free(current, false); |
a6f76f23d
|
1748 1749 1750 |
if (displaced) put_files_struct(displaced); return retval; |
1da177e4c
|
1751 |
|
a6f76f23d
|
1752 |
out: |
8890b2934
|
1753 1754 1755 1756 1757 1758 1759 1760 |
/* * If past the point of no return ensure the the code never * returns to the userspace process. Use an existing fatal * signal if present otherwise terminate the process with * SIGSEGV. */ if (bprm->point_of_no_return && !fatal_signal_pending(current)) force_sigsegv(SIGSEGV); |
1da177e4c
|
1761 |
|
498052bba
|
1762 |
out_unmark: |
9e00cdb09
|
1763 |
current->fs->in_exec = 0; |
f9ce1f1cd
|
1764 |
current->in_execve = 0; |
a6f76f23d
|
1765 |
|
0a8f36eb4
|
1766 1767 1768 |
out_files: if (displaced) reset_files_struct(displaced); |
0c9cdff05
|
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 1797 1798 1799 1800 1801 1802 1803 1804 |
return retval; } static int do_execveat_common(int fd, struct filename *filename, struct user_arg_ptr argv, struct user_arg_ptr envp, int flags) { struct linux_binprm *bprm; int retval; if (IS_ERR(filename)) return PTR_ERR(filename); /* * 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) && atomic_read(¤t_user()->processes) > rlimit(RLIMIT_NPROC)) { 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; bprm = alloc_bprm(fd, filename); if (IS_ERR(bprm)) { retval = PTR_ERR(bprm); goto out_ret; } |
d8b9cd549
|
1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 |
retval = count(argv, MAX_ARG_STRINGS); if (retval < 0) goto out_free; bprm->argc = retval; retval = count(envp, MAX_ARG_STRINGS); if (retval < 0) goto out_free; bprm->envc = retval; retval = bprm_stack_limits(bprm); |
0c9cdff05
|
1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 |
if (retval < 0) goto out_free; retval = copy_string_kernel(bprm->filename, bprm); if (retval < 0) goto out_free; bprm->exec = bprm->p; retval = copy_strings(bprm->envc, envp, bprm); if (retval < 0) goto out_free; retval = copy_strings(bprm->argc, argv, bprm); if (retval < 0) goto out_free; retval = bprm_execve(bprm, fd, filename, flags); |
a6f76f23d
|
1833 |
out_free: |
08a6fac1c
|
1834 |
free_bprm(bprm); |
1da177e4c
|
1835 1836 |
out_ret: |
25cf336de
|
1837 |
putname(filename); |
1da177e4c
|
1838 1839 |
return retval; } |
be619f7f0
|
1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 |
int kernel_execve(const char *kernel_filename, const char *const *argv, const char *const *envp) { struct filename *filename; struct linux_binprm *bprm; int fd = AT_FDCWD; int retval; filename = getname_kernel(kernel_filename); if (IS_ERR(filename)) return PTR_ERR(filename); bprm = alloc_bprm(fd, filename); if (IS_ERR(bprm)) { retval = PTR_ERR(bprm); goto out_ret; } retval = count_strings_kernel(argv); if (retval < 0) goto out_free; bprm->argc = retval; retval = count_strings_kernel(envp); if (retval < 0) goto out_free; bprm->envc = retval; retval = bprm_stack_limits(bprm); if (retval < 0) goto out_free; retval = copy_string_kernel(bprm->filename, bprm); if (retval < 0) goto out_free; bprm->exec = bprm->p; retval = copy_strings_kernel(bprm->envc, envp, bprm); if (retval < 0) goto out_free; retval = copy_strings_kernel(bprm->argc, argv, bprm); if (retval < 0) goto out_free; retval = bprm_execve(bprm, fd, filename, 0); out_free: free_bprm(bprm); out_ret: putname(filename); return retval; } static int do_execve(struct filename *filename, |
ba2d01629
|
1894 |
const char __user *const __user *__argv, |
da3d4c5fa
|
1895 |
const char __user *const __user *__envp) |
ba2d01629
|
1896 |
{ |
0e028465d
|
1897 1898 |
struct user_arg_ptr argv = { .ptr.native = __argv }; struct user_arg_ptr envp = { .ptr.native = __envp }; |
51f39a1f0
|
1899 1900 |
return do_execveat_common(AT_FDCWD, filename, argv, envp, 0); } |
be619f7f0
|
1901 |
static int do_execveat(int fd, struct filename *filename, |
51f39a1f0
|
1902 1903 1904 1905 1906 1907 1908 1909 |
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
|
1910 1911 1912 |
} #ifdef CONFIG_COMPAT |
c4ad8f98b
|
1913 |
static int compat_do_execve(struct filename *filename, |
38b983b34
|
1914 |
const compat_uptr_t __user *__argv, |
d03d26e58
|
1915 |
const compat_uptr_t __user *__envp) |
0e028465d
|
1916 1917 1918 1919 1920 1921 1922 1923 1924 |
{ struct user_arg_ptr argv = { .is_compat = true, .ptr.compat = __argv, }; struct user_arg_ptr envp = { .is_compat = true, .ptr.compat = __envp, }; |
51f39a1f0
|
1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 |
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
|
1942 |
} |
0e028465d
|
1943 |
#endif |
ba2d01629
|
1944 |
|
964ee7df9
|
1945 |
void set_binfmt(struct linux_binfmt *new) |
1da177e4c
|
1946 |
{ |
801460d0c
|
1947 1948 1949 1950 |
struct mm_struct *mm = current->mm; if (mm->binfmt) module_put(mm->binfmt->module); |
1da177e4c
|
1951 |
|
801460d0c
|
1952 |
mm->binfmt = new; |
964ee7df9
|
1953 1954 |
if (new) __module_get(new->module); |
1da177e4c
|
1955 |
} |
1da177e4c
|
1956 |
EXPORT_SYMBOL(set_binfmt); |
6c5d52382
|
1957 |
/* |
7288e1187
|
1958 |
* set_dumpable stores three-value SUID_DUMP_* into mm->flags. |
6c5d52382
|
1959 1960 1961 |
*/ void set_dumpable(struct mm_struct *mm, int value) { |
7288e1187
|
1962 1963 |
if (WARN_ON((unsigned)value > SUID_DUMP_ROOT)) return; |
26e152252
|
1964 |
set_mask_bits(&mm->flags, MMF_DUMPABLE_MASK, value); |
6c5d52382
|
1965 |
} |
6c5d52382
|
1966 |
|
38b983b34
|
1967 1968 1969 1970 1971 |
SYSCALL_DEFINE3(execve, const char __user *, filename, const char __user *const __user *, argv, const char __user *const __user *, envp) { |
c4ad8f98b
|
1972 |
return do_execve(getname(filename), argv, envp); |
38b983b34
|
1973 |
} |
51f39a1f0
|
1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 |
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
|
1987 |
#ifdef CONFIG_COMPAT |
625b1d7e8
|
1988 1989 1990 |
COMPAT_SYSCALL_DEFINE3(execve, const char __user *, filename, const compat_uptr_t __user *, argv, const compat_uptr_t __user *, envp) |
38b983b34
|
1991 |
{ |
c4ad8f98b
|
1992 |
return compat_do_execve(getname(filename), argv, envp); |
38b983b34
|
1993 |
} |
51f39a1f0
|
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 |
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
|
2007 |
#endif |