Blame view
fs/exec.c
44.3 KB
1da177e4c
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 |
/* * 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 |
c1df5a637
|
22 |
* formats. |
1da177e4c
|
23 |
*/ |
1da177e4c
|
24 25 |
#include <linux/slab.h> #include <linux/file.h> |
9f3acc314
|
26 |
#include <linux/fdtable.h> |
ba92a43db
|
27 |
#include <linux/mm.h> |
615d6e875
|
28 |
#include <linux/vmacache.h> |
1da177e4c
|
29 30 |
#include <linux/stat.h> #include <linux/fcntl.h> |
ba92a43db
|
31 |
#include <linux/swap.h> |
74aadce98
|
32 |
#include <linux/string.h> |
1da177e4c
|
33 |
#include <linux/init.h> |
ca5b172bd
|
34 |
#include <linux/pagemap.h> |
cdd6c482c
|
35 |
#include <linux/perf_event.h> |
1da177e4c
|
36 37 38 39 40 |
#include <linux/highmem.h> #include <linux/spinlock.h> #include <linux/key.h> #include <linux/personality.h> #include <linux/binfmts.h> |
1da177e4c
|
41 |
#include <linux/utsname.h> |
84d737866
|
42 |
#include <linux/pid_namespace.h> |
1da177e4c
|
43 44 |
#include <linux/module.h> #include <linux/namei.h> |
1da177e4c
|
45 46 47 |
#include <linux/mount.h> #include <linux/security.h> #include <linux/syscalls.h> |
8f0ab5147
|
48 |
#include <linux/tsacct_kern.h> |
9f46080c4
|
49 |
#include <linux/cn_proc.h> |
473ae30bc
|
50 |
#include <linux/audit.h> |
6341c393f
|
51 |
#include <linux/tracehook.h> |
5f4123be3
|
52 |
#include <linux/kmod.h> |
6110e3abb
|
53 |
#include <linux/fsnotify.h> |
5ad4e53bd
|
54 |
#include <linux/fs_struct.h> |
61be228a0
|
55 |
#include <linux/pipe_fs_i.h> |
3d5992d2a
|
56 |
#include <linux/oom.h> |
0e028465d
|
57 |
#include <linux/compat.h> |
b44a7dfc6
|
58 |
#include <linux/vmalloc.h> |
1da177e4c
|
59 60 61 |
#include <asm/uaccess.h> #include <asm/mmu_context.h> |
b6a2fea39
|
62 |
#include <asm/tlb.h> |
43d2b1132
|
63 64 |
#include <trace/events/task.h> |
a6f76f23d
|
65 |
#include "internal.h" |
1da177e4c
|
66 |
|
4ff16c25e
|
67 |
#include <trace/events/sched.h> |
d6e711448
|
68 |
int suid_dumpable = 0; |
e4dc1b14d
|
69 |
static LIST_HEAD(formats); |
1da177e4c
|
70 |
static DEFINE_RWLOCK(binfmt_lock); |
8fc3dc5a3
|
71 |
void __register_binfmt(struct linux_binfmt * fmt, int insert) |
1da177e4c
|
72 |
{ |
8fc3dc5a3
|
73 |
BUG_ON(!fmt); |
92eaa565a
|
74 75 |
if (WARN_ON(!fmt->load_binary)) return; |
1da177e4c
|
76 |
write_lock(&binfmt_lock); |
74641f584
|
77 78 |
insert ? list_add(&fmt->lh, &formats) : list_add_tail(&fmt->lh, &formats); |
1da177e4c
|
79 |
write_unlock(&binfmt_lock); |
1da177e4c
|
80 |
} |
74641f584
|
81 |
EXPORT_SYMBOL(__register_binfmt); |
1da177e4c
|
82 |
|
f6b450d48
|
83 |
void unregister_binfmt(struct linux_binfmt * fmt) |
1da177e4c
|
84 |
{ |
1da177e4c
|
85 |
write_lock(&binfmt_lock); |
e4dc1b14d
|
86 |
list_del(&fmt->lh); |
1da177e4c
|
87 |
write_unlock(&binfmt_lock); |
1da177e4c
|
88 89 90 91 92 93 94 95 |
} EXPORT_SYMBOL(unregister_binfmt); static inline void put_binfmt(struct linux_binfmt * fmt) { module_put(fmt->module); } |
90f8572b0
|
96 97 98 99 100 |
bool path_noexec(const struct path *path) { return (path->mnt->mnt_flags & MNT_NOEXEC) || (path->mnt->mnt_sb->s_iflags & SB_I_NOEXEC); } |
69369a700
|
101 |
#ifdef CONFIG_USELIB |
1da177e4c
|
102 103 104 105 106 107 |
/* * 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
|
108 |
SYSCALL_DEFINE1(uselib, const char __user *, library) |
1da177e4c
|
109 |
{ |
72c2d5319
|
110 |
struct linux_binfmt *fmt; |
964bd1836
|
111 |
struct file *file; |
91a27b2a7
|
112 |
struct filename *tmp = getname(library); |
964bd1836
|
113 |
int error = PTR_ERR(tmp); |
47c805dc2
|
114 115 |
static const struct open_flags uselib_flags = { .open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, |
62fb4a155
|
116 |
.acc_mode = MAY_READ | MAY_EXEC, |
f9652e10c
|
117 118 |
.intent = LOOKUP_OPEN, .lookup_flags = LOOKUP_FOLLOW, |
47c805dc2
|
119 |
}; |
964bd1836
|
120 |
|
6e8341a11
|
121 122 |
if (IS_ERR(tmp)) goto out; |
f9652e10c
|
123 |
file = do_filp_open(AT_FDCWD, tmp, &uselib_flags); |
6e8341a11
|
124 125 126 |
putname(tmp); error = PTR_ERR(file); if (IS_ERR(file)) |
1da177e4c
|
127 128 129 |
goto out; error = -EINVAL; |
496ad9aa8
|
130 |
if (!S_ISREG(file_inode(file)->i_mode)) |
1da177e4c
|
131 |
goto exit; |
30524472c
|
132 |
error = -EACCES; |
90f8572b0
|
133 |
if (path_noexec(&file->f_path)) |
6146f0d5e
|
134 |
goto exit; |
1da177e4c
|
135 |
|
2a12a9d78
|
136 |
fsnotify_open(file); |
6110e3abb
|
137 |
|
1da177e4c
|
138 |
error = -ENOEXEC; |
1da177e4c
|
139 |
|
72c2d5319
|
140 141 142 143 144 145 |
read_lock(&binfmt_lock); list_for_each_entry(fmt, &formats, lh) { if (!fmt->load_shlib) continue; if (!try_module_get(fmt->module)) continue; |
1da177e4c
|
146 |
read_unlock(&binfmt_lock); |
72c2d5319
|
147 148 149 150 151 |
error = fmt->load_shlib(file); read_lock(&binfmt_lock); put_binfmt(fmt); if (error != -ENOEXEC) break; |
1da177e4c
|
152 |
} |
72c2d5319
|
153 |
read_unlock(&binfmt_lock); |
6e8341a11
|
154 |
exit: |
1da177e4c
|
155 156 157 |
fput(file); out: return error; |
1da177e4c
|
158 |
} |
69369a700
|
159 |
#endif /* #ifdef CONFIG_USELIB */ |
1da177e4c
|
160 |
|
b6a2fea39
|
161 |
#ifdef CONFIG_MMU |
ae6b585ee
|
162 163 164 165 166 167 |
/* * 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
|
168 |
static void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) |
3c77f8457
|
169 170 171 172 173 174 175 176 |
{ struct mm_struct *mm = current->mm; long diff = (long)(pages - bprm->vma_pages); if (!mm || !diff) return; bprm->vma_pages = pages; |
3c77f8457
|
177 |
add_mm_counter(mm, MM_ANONPAGES, diff); |
3c77f8457
|
178 |
} |
0e028465d
|
179 |
static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, |
b6a2fea39
|
180 181 182 183 |
int write) { struct page *page; int ret; |
9beae1ea8
|
184 |
unsigned int gup_flags = FOLL_FORCE; |
b6a2fea39
|
185 186 187 |
#ifdef CONFIG_STACK_GROWSUP if (write) { |
d05f3169c
|
188 |
ret = expand_downwards(bprm->vma, pos); |
b6a2fea39
|
189 190 191 192 |
if (ret < 0) return NULL; } #endif |
9beae1ea8
|
193 194 195 |
if (write) gup_flags |= FOLL_WRITE; |
1e9877902
|
196 197 198 199 |
/* * We are doing an exec(). 'current' is the process * doing the exec and bprm->mm is the new process's mm. */ |
9beae1ea8
|
200 201 |
ret = get_user_pages_remote(current, bprm->mm, pos, 1, gup_flags, &page, NULL); |
b6a2fea39
|
202 203 204 205 |
if (ret <= 0) return NULL; if (write) { |
b6a2fea39
|
206 |
unsigned long size = bprm->vma->vm_end - bprm->vma->vm_start; |
a64e715fc
|
207 |
struct rlimit *rlim; |
3c77f8457
|
208 |
acct_arg_size(bprm, size / PAGE_SIZE); |
a64e715fc
|
209 210 211 212 213 214 |
/* * We've historically supported up to 32 pages (ARG_MAX) * of argument strings even with small stacks */ if (size <= ARG_MAX) return page; |
b6a2fea39
|
215 216 217 218 219 220 221 222 |
/* * Limit to 1/4-th the stack size for the argv+env strings. * This ensures that: * - the remaining binfmt code will not run out of stack space, * - the program will have a reasonable amount of stack left * to work from. */ |
a64e715fc
|
223 |
rlim = current->signal->rlim; |
d554ed895
|
224 |
if (size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur) / 4) { |
b6a2fea39
|
225 226 227 228 229 230 231 232 233 234 235 236 |
put_page(page); return NULL; } } return page; } static void put_arg_page(struct page *page) { put_page(page); } |
b6a2fea39
|
237 238 239 240 241 242 243 244 245 246 247 248 |
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
|
249 |
int err; |
b6a2fea39
|
250 251 252 253 254 |
struct vm_area_struct *vma = NULL; struct mm_struct *mm = bprm->mm; bprm->vma = vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (!vma) |
eaccbfa56
|
255 |
return -ENOMEM; |
b6a2fea39
|
256 |
|
f268dfe90
|
257 258 259 260 |
if (down_write_killable(&mm->mmap_sem)) { err = -EINTR; goto err_free; } |
b6a2fea39
|
261 262 263 264 265 266 267 268 |
vma->vm_mm = mm; /* * Place the stack at the largest stack address the architecture * supports. Later, we'll move this to an appropriate place. We don't * use STACK_TOP because that can depend on attributes which aren't * configured yet. */ |
aacb3d17a
|
269 |
BUILD_BUG_ON(VM_STACK_FLAGS & VM_STACK_INCOMPLETE_SETUP); |
b6a2fea39
|
270 271 |
vma->vm_end = STACK_TOP_MAX; vma->vm_start = vma->vm_end - PAGE_SIZE; |
d9104d1ca
|
272 |
vma->vm_flags = VM_SOFTDIRTY | VM_STACK_FLAGS | VM_STACK_INCOMPLETE_SETUP; |
3ed75eb8f
|
273 |
vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); |
5beb49305
|
274 |
INIT_LIST_HEAD(&vma->anon_vma_chain); |
462e635e5
|
275 |
|
b6a2fea39
|
276 |
err = insert_vm_struct(mm, vma); |
eaccbfa56
|
277 |
if (err) |
b6a2fea39
|
278 |
goto err; |
b6a2fea39
|
279 280 |
mm->stack_vm = mm->total_vm = 1; |
fe3d197f8
|
281 |
arch_bprm_mm_init(mm, vma); |
b6a2fea39
|
282 |
up_write(&mm->mmap_sem); |
b6a2fea39
|
283 |
bprm->p = vma->vm_end - sizeof(void *); |
b6a2fea39
|
284 |
return 0; |
b6a2fea39
|
285 |
err: |
eaccbfa56
|
286 |
up_write(&mm->mmap_sem); |
f268dfe90
|
287 |
err_free: |
eaccbfa56
|
288 289 |
bprm->vma = NULL; kmem_cache_free(vm_area_cachep, vma); |
b6a2fea39
|
290 291 292 293 294 295 296 297 298 |
return err; } static bool valid_arg_len(struct linux_binprm *bprm, long len) { return len <= MAX_ARG_STRLEN; } #else |
0e028465d
|
299 |
static inline void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) |
3c77f8457
|
300 301 |
{ } |
0e028465d
|
302 |
static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, |
b6a2fea39
|
303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 |
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
|
362 |
static int bprm_mm_init(struct linux_binprm *bprm) |
b6a2fea39
|
363 364 365 366 367 368 369 370 |
{ int err; struct mm_struct *mm = NULL; bprm->mm = mm = mm_alloc(); err = -ENOMEM; if (!mm) goto err; |
b6a2fea39
|
371 372 373 374 375 376 377 378 379 380 381 382 383 384 |
err = __bprm_mm_init(bprm); if (err) goto err; return 0; err: if (mm) { bprm->mm = NULL; mmdrop(mm); } return err; } |
ba2d01629
|
385 |
struct user_arg_ptr { |
0e028465d
|
386 387 388 389 390 391 |
#ifdef CONFIG_COMPAT bool is_compat; #endif union { const char __user *const __user *native; #ifdef CONFIG_COMPAT |
38b983b34
|
392 |
const compat_uptr_t __user *compat; |
0e028465d
|
393 394 |
#endif } ptr; |
ba2d01629
|
395 396 397 |
}; static const char __user *get_user_arg_ptr(struct user_arg_ptr argv, int nr) |
1d1dbf813
|
398 |
{ |
0e028465d
|
399 400 401 402 403 404 405 406 |
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
|
407 |
|
0e028465d
|
408 409 410 411 412 |
return compat_ptr(compat); } #endif if (get_user(native, argv.ptr.native + nr)) |
1d1dbf813
|
413 |
return ERR_PTR(-EFAULT); |
0e028465d
|
414 |
return native; |
1d1dbf813
|
415 |
} |
1da177e4c
|
416 417 418 |
/* * count() counts the number of strings in array ARGV. */ |
ba2d01629
|
419 |
static int count(struct user_arg_ptr argv, int max) |
1da177e4c
|
420 421 |
{ int i = 0; |
0e028465d
|
422 |
if (argv.ptr.native != NULL) { |
1da177e4c
|
423 |
for (;;) { |
1d1dbf813
|
424 |
const char __user *p = get_user_arg_ptr(argv, i); |
1da177e4c
|
425 |
|
1da177e4c
|
426 427 |
if (!p) break; |
1d1dbf813
|
428 429 430 |
if (IS_ERR(p)) return -EFAULT; |
6d92d4f6a
|
431 |
if (i >= max) |
1da177e4c
|
432 |
return -E2BIG; |
6d92d4f6a
|
433 |
++i; |
9aea5a65a
|
434 435 436 |
if (fatal_signal_pending(current)) return -ERESTARTNOHAND; |
1da177e4c
|
437 438 439 440 441 442 443 |
cond_resched(); } } return i; } /* |
b6a2fea39
|
444 445 446 |
* '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
|
447 |
*/ |
ba2d01629
|
448 |
static int copy_strings(int argc, struct user_arg_ptr argv, |
75c96f858
|
449 |
struct linux_binprm *bprm) |
1da177e4c
|
450 451 452 |
{ struct page *kmapped_page = NULL; char *kaddr = NULL; |
b6a2fea39
|
453 |
unsigned long kpos = 0; |
1da177e4c
|
454 455 456 |
int ret; while (argc-- > 0) { |
d7627467b
|
457 |
const char __user *str; |
1da177e4c
|
458 459 |
int len; unsigned long pos; |
1d1dbf813
|
460 461 462 |
ret = -EFAULT; str = get_user_arg_ptr(argv, argc); if (IS_ERR(str)) |
1da177e4c
|
463 |
goto out; |
1da177e4c
|
464 |
|
1d1dbf813
|
465 466 467 468 469 470 |
len = strnlen_user(str, MAX_ARG_STRLEN); if (!len) goto out; ret = -E2BIG; if (!valid_arg_len(bprm, len)) |
1da177e4c
|
471 |
goto out; |
1da177e4c
|
472 |
|
b6a2fea39
|
473 |
/* We're going to work our way backwords. */ |
1da177e4c
|
474 |
pos = bprm->p; |
b6a2fea39
|
475 476 |
str += len; bprm->p -= len; |
1da177e4c
|
477 478 |
while (len > 0) { |
1da177e4c
|
479 |
int offset, bytes_to_copy; |
1da177e4c
|
480 |
|
9aea5a65a
|
481 482 483 484 |
if (fatal_signal_pending(current)) { ret = -ERESTARTNOHAND; goto out; } |
7993bc1f4
|
485 |
cond_resched(); |
1da177e4c
|
486 |
offset = pos % PAGE_SIZE; |
b6a2fea39
|
487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 |
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
|
503 |
if (!page) { |
b6a2fea39
|
504 |
ret = -E2BIG; |
1da177e4c
|
505 506 |
goto out; } |
1da177e4c
|
507 |
|
b6a2fea39
|
508 509 |
if (kmapped_page) { flush_kernel_dcache_page(kmapped_page); |
1da177e4c
|
510 |
kunmap(kmapped_page); |
b6a2fea39
|
511 512 |
put_arg_page(kmapped_page); } |
1da177e4c
|
513 514 |
kmapped_page = page; kaddr = kmap(kmapped_page); |
b6a2fea39
|
515 516 |
kpos = pos & PAGE_MASK; flush_arg_page(bprm, kpos, kmapped_page); |
1da177e4c
|
517 |
} |
b6a2fea39
|
518 |
if (copy_from_user(kaddr+offset, str, bytes_to_copy)) { |
1da177e4c
|
519 520 521 |
ret = -EFAULT; goto out; } |
1da177e4c
|
522 523 524 525 |
} } ret = 0; out: |
b6a2fea39
|
526 527 |
if (kmapped_page) { flush_kernel_dcache_page(kmapped_page); |
1da177e4c
|
528 |
kunmap(kmapped_page); |
b6a2fea39
|
529 530 |
put_arg_page(kmapped_page); } |
1da177e4c
|
531 532 533 534 535 536 |
return ret; } /* * Like copy_strings, but get argv and its values from kernel memory. */ |
ba2d01629
|
537 |
int copy_strings_kernel(int argc, const char *const *__argv, |
d7627467b
|
538 |
struct linux_binprm *bprm) |
1da177e4c
|
539 540 541 |
{ int r; mm_segment_t oldfs = get_fs(); |
ba2d01629
|
542 |
struct user_arg_ptr argv = { |
0e028465d
|
543 |
.ptr.native = (const char __user *const __user *)__argv, |
ba2d01629
|
544 |
}; |
1da177e4c
|
545 |
set_fs(KERNEL_DS); |
ba2d01629
|
546 |
r = copy_strings(argc, argv, bprm); |
1da177e4c
|
547 |
set_fs(oldfs); |
ba2d01629
|
548 |
|
1da177e4c
|
549 550 |
return r; } |
1da177e4c
|
551 552 553 |
EXPORT_SYMBOL(copy_strings_kernel); #ifdef CONFIG_MMU |
b6a2fea39
|
554 |
|
1da177e4c
|
555 |
/* |
b6a2fea39
|
556 557 558 |
* 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
|
559 |
* |
b6a2fea39
|
560 561 562 563 564 565 |
* 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
|
566 |
*/ |
b6a2fea39
|
567 |
static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) |
1da177e4c
|
568 569 |
{ struct mm_struct *mm = vma->vm_mm; |
b6a2fea39
|
570 571 572 573 574 |
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
|
575 |
struct mmu_gather tlb; |
1da177e4c
|
576 |
|
b6a2fea39
|
577 |
BUG_ON(new_start > new_end); |
1da177e4c
|
578 |
|
b6a2fea39
|
579 580 581 582 583 584 585 586 587 588 |
/* * 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
|
589 590 |
if (vma_adjust(vma, new_start, old_end, vma->vm_pgoff, NULL)) return -ENOMEM; |
b6a2fea39
|
591 592 593 594 595 596 |
/* * 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
|
597 |
vma, new_start, length, false)) |
b6a2fea39
|
598 599 600 |
return -ENOMEM; lru_add_drain(); |
2b047252d
|
601 |
tlb_gather_mmu(&tlb, mm, old_start, old_end); |
b6a2fea39
|
602 603 604 605 |
if (new_end > old_start) { /* * when the old and new regions overlap clear from new_end. */ |
d16dfc550
|
606 |
free_pgd_range(&tlb, new_end, old_end, new_end, |
6ee8630e0
|
607 |
vma->vm_next ? vma->vm_next->vm_start : USER_PGTABLES_CEILING); |
b6a2fea39
|
608 609 610 611 612 613 614 |
} 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
|
615 |
free_pgd_range(&tlb, old_start, old_end, new_end, |
6ee8630e0
|
616 |
vma->vm_next ? vma->vm_next->vm_start : USER_PGTABLES_CEILING); |
1da177e4c
|
617 |
} |
2b047252d
|
618 |
tlb_finish_mmu(&tlb, old_start, old_end); |
b6a2fea39
|
619 620 |
/* |
5beb49305
|
621 |
* Shrink the vma to just the new range. Always succeeds. |
b6a2fea39
|
622 623 624 625 |
*/ vma_adjust(vma, new_start, new_end, vma->vm_pgoff, NULL); return 0; |
1da177e4c
|
626 |
} |
b6a2fea39
|
627 628 629 630 |
/* * Finalizes the stack vm_area_struct. The flags and permissions are updated, * the stack is optionally relocated, and some extra space is added. */ |
1da177e4c
|
631 632 633 634 |
int setup_arg_pages(struct linux_binprm *bprm, unsigned long stack_top, int executable_stack) { |
b6a2fea39
|
635 636 |
unsigned long ret; unsigned long stack_shift; |
1da177e4c
|
637 |
struct mm_struct *mm = current->mm; |
b6a2fea39
|
638 639 640 641 |
struct vm_area_struct *vma = bprm->vma; struct vm_area_struct *prev = NULL; unsigned long vm_flags; unsigned long stack_base; |
803bf5ec2
|
642 643 644 |
unsigned long stack_size; unsigned long stack_expand; unsigned long rlim_stack; |
1da177e4c
|
645 646 |
#ifdef CONFIG_STACK_GROWSUP |
d71f290b4
|
647 |
/* Limit stack size */ |
d554ed895
|
648 |
stack_base = rlimit_max(RLIMIT_STACK); |
d71f290b4
|
649 650 |
if (stack_base > STACK_SIZE_MAX) stack_base = STACK_SIZE_MAX; |
1da177e4c
|
651 |
|
d045c77c1
|
652 653 |
/* Add space for stack randomization. */ stack_base += (STACK_RND_MASK << PAGE_SHIFT); |
b6a2fea39
|
654 655 656 |
/* Make sure we didn't let the argument array grow too large. */ if (vma->vm_end - vma->vm_start > stack_base) return -ENOMEM; |
1da177e4c
|
657 |
|
b6a2fea39
|
658 |
stack_base = PAGE_ALIGN(stack_top - stack_base); |
1da177e4c
|
659 |
|
b6a2fea39
|
660 661 662 |
stack_shift = vma->vm_start - stack_base; mm->arg_start = bprm->p - stack_shift; bprm->p = vma->vm_end - stack_shift; |
1da177e4c
|
663 |
#else |
b6a2fea39
|
664 665 |
stack_top = arch_align_stack(stack_top); stack_top = PAGE_ALIGN(stack_top); |
1b528181b
|
666 667 668 669 |
if (unlikely(stack_top < mmap_min_addr) || unlikely(vma->vm_end - vma->vm_start >= stack_top - mmap_min_addr)) return -ENOMEM; |
b6a2fea39
|
670 671 672 |
stack_shift = vma->vm_end - stack_top; bprm->p -= stack_shift; |
1da177e4c
|
673 |
mm->arg_start = bprm->p; |
1da177e4c
|
674 |
#endif |
1da177e4c
|
675 |
if (bprm->loader) |
b6a2fea39
|
676 677 |
bprm->loader -= stack_shift; bprm->exec -= stack_shift; |
1da177e4c
|
678 |
|
f268dfe90
|
679 680 |
if (down_write_killable(&mm->mmap_sem)) return -EINTR; |
96a8e13ed
|
681 |
vm_flags = VM_STACK_FLAGS; |
b6a2fea39
|
682 683 684 685 686 687 688 689 690 691 692 |
/* * 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
|
693 |
vm_flags |= VM_STACK_INCOMPLETE_SETUP; |
b6a2fea39
|
694 695 696 697 698 699 700 701 702 703 |
ret = mprotect_fixup(vma, &prev, vma->vm_start, vma->vm_end, vm_flags); if (ret) goto out_unlock; BUG_ON(prev != vma); /* Move stack pages down in memory. */ if (stack_shift) { ret = shift_arg_pages(vma, stack_shift); |
fc63cf237
|
704 705 |
if (ret) goto out_unlock; |
1da177e4c
|
706 |
} |
a8bef8ff6
|
707 708 |
/* mprotect_fixup is overkill to remove the temporary stack flags */ vma->vm_flags &= ~VM_STACK_INCOMPLETE_SETUP; |
5ef097dd7
|
709 |
stack_expand = 131072UL; /* randomly 32*4k (or 2*64k) pages */ |
803bf5ec2
|
710 711 712 713 714 715 |
stack_size = vma->vm_end - vma->vm_start; /* * Align this down to a page boundary as expand_stack * will align it up. */ rlim_stack = rlimit(RLIMIT_STACK) & PAGE_MASK; |
b6a2fea39
|
716 |
#ifdef CONFIG_STACK_GROWSUP |
803bf5ec2
|
717 718 719 720 |
if (stack_size + stack_expand > rlim_stack) stack_base = vma->vm_start + rlim_stack; else stack_base = vma->vm_end + stack_expand; |
b6a2fea39
|
721 |
#else |
803bf5ec2
|
722 723 724 725 |
if (stack_size + stack_expand > rlim_stack) stack_base = vma->vm_end - rlim_stack; else stack_base = vma->vm_start - stack_expand; |
b6a2fea39
|
726 |
#endif |
3af9e8592
|
727 |
current->mm->start_stack = bprm->p; |
b6a2fea39
|
728 729 730 731 732 |
ret = expand_stack(vma, stack_base); if (ret) ret = -EFAULT; out_unlock: |
1da177e4c
|
733 |
up_write(&mm->mmap_sem); |
fc63cf237
|
734 |
return ret; |
1da177e4c
|
735 |
} |
1da177e4c
|
736 |
EXPORT_SYMBOL(setup_arg_pages); |
7e7ec6a93
|
737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 |
#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
|
769 |
#endif /* CONFIG_MMU */ |
51f39a1f0
|
770 |
static struct file *do_open_execat(int fd, struct filename *name, int flags) |
1da177e4c
|
771 |
{ |
1da177e4c
|
772 |
struct file *file; |
e56b6a5dd
|
773 |
int err; |
51f39a1f0
|
774 |
struct open_flags open_exec_flags = { |
47c805dc2
|
775 |
.open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, |
62fb4a155
|
776 |
.acc_mode = MAY_EXEC, |
f9652e10c
|
777 778 |
.intent = LOOKUP_OPEN, .lookup_flags = LOOKUP_FOLLOW, |
47c805dc2
|
779 |
}; |
1da177e4c
|
780 |
|
51f39a1f0
|
781 782 783 784 785 786 787 788 |
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
|
789 |
if (IS_ERR(file)) |
e56b6a5dd
|
790 791 792 |
goto out; err = -EACCES; |
496ad9aa8
|
793 |
if (!S_ISREG(file_inode(file)->i_mode)) |
6e8341a11
|
794 |
goto exit; |
e56b6a5dd
|
795 |
|
90f8572b0
|
796 |
if (path_noexec(&file->f_path)) |
6e8341a11
|
797 |
goto exit; |
e56b6a5dd
|
798 799 |
err = deny_write_access(file); |
6e8341a11
|
800 801 |
if (err) goto exit; |
1da177e4c
|
802 |
|
51f39a1f0
|
803 804 |
if (name->name[0] != '\0') fsnotify_open(file); |
6e8341a11
|
805 |
out: |
e56b6a5dd
|
806 |
return file; |
6e8341a11
|
807 808 |
exit: fput(file); |
e56b6a5dd
|
809 810 |
return ERR_PTR(err); } |
c4ad8f98b
|
811 812 813 |
struct file *open_exec(const char *name) { |
516891041
|
814 815 816 817 818 819 820 821 |
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
|
822 |
} |
1da177e4c
|
823 |
EXPORT_SYMBOL(open_exec); |
6777d773a
|
824 825 |
int kernel_read(struct file *file, loff_t offset, char *addr, unsigned long count) |
1da177e4c
|
826 827 828 829 830 831 832 833 834 835 836 837 838 839 |
{ mm_segment_t old_fs; loff_t pos = offset; int result; old_fs = get_fs(); set_fs(get_ds()); /* The cast to a user pointer is valid due to the set_fs() */ result = vfs_read(file, (void __user *)addr, count, &pos); set_fs(old_fs); return result; } EXPORT_SYMBOL(kernel_read); |
b44a7dfc6
|
840 |
int kernel_read_file(struct file *file, void **buf, loff_t *size, |
bc8ca5b92
|
841 |
loff_t max_size, enum kernel_read_file_id id) |
b44a7dfc6
|
842 843 844 845 846 847 848 |
{ loff_t i_size, pos; ssize_t bytes = 0; int ret; if (!S_ISREG(file_inode(file)->i_mode) || max_size < 0) return -EINVAL; |
39eeb4fb9
|
849 850 851 |
ret = security_kernel_read_file(file, id); if (ret) return ret; |
39d637af5
|
852 853 854 |
ret = deny_write_access(file); if (ret) return ret; |
b44a7dfc6
|
855 |
i_size = i_size_read(file_inode(file)); |
39d637af5
|
856 857 858 859 860 861 862 863 |
if (max_size > 0 && i_size > max_size) { ret = -EFBIG; goto out; } if (i_size <= 0) { ret = -EINVAL; goto out; } |
b44a7dfc6
|
864 |
|
a098ecd2f
|
865 866 |
if (id != READING_FIRMWARE_PREALLOC_BUFFER) *buf = vmalloc(i_size); |
39d637af5
|
867 868 869 870 |
if (!*buf) { ret = -ENOMEM; goto out; } |
b44a7dfc6
|
871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 |
pos = 0; while (pos < i_size) { bytes = kernel_read(file, pos, (char *)(*buf) + pos, i_size - pos); if (bytes < 0) { ret = bytes; goto out; } if (bytes == 0) break; pos += bytes; } if (pos != i_size) { ret = -EIO; |
39d637af5
|
888 |
goto out_free; |
b44a7dfc6
|
889 |
} |
bc8ca5b92
|
890 |
ret = security_kernel_post_read_file(file, *buf, i_size, id); |
b44a7dfc6
|
891 892 |
if (!ret) *size = pos; |
39d637af5
|
893 |
out_free: |
b44a7dfc6
|
894 |
if (ret < 0) { |
a098ecd2f
|
895 896 897 898 |
if (id != READING_FIRMWARE_PREALLOC_BUFFER) { vfree(*buf); *buf = NULL; } |
b44a7dfc6
|
899 |
} |
39d637af5
|
900 901 902 |
out: allow_write_access(file); |
b44a7dfc6
|
903 904 905 |
return ret; } EXPORT_SYMBOL_GPL(kernel_read_file); |
09596b94f
|
906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 |
int kernel_read_file_from_path(char *path, void **buf, loff_t *size, 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
|
924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 |
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
|
939 940 |
ssize_t read_code(struct file *file, unsigned long addr, loff_t pos, size_t len) { |
ec6955798
|
941 |
ssize_t res = vfs_read(file, (void __user *)addr, len, &pos); |
3dc20cb28
|
942 943 944 945 946 |
if (res > 0) flush_icache_range(addr, addr + len); return res; } EXPORT_SYMBOL(read_code); |
1da177e4c
|
947 948 949 |
static int exec_mmap(struct mm_struct *mm) { struct task_struct *tsk; |
615d6e875
|
950 |
struct mm_struct *old_mm, *active_mm; |
1da177e4c
|
951 952 953 954 955 956 957 |
/* Notify parent that we're no longer interested in the old VM */ tsk = current; old_mm = current->mm; mm_release(tsk, old_mm); if (old_mm) { |
4fe7efdbd
|
958 |
sync_mm_rss(old_mm); |
1da177e4c
|
959 960 961 962 |
/* * 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
|
963 |
* checking core_state and changing tsk->mm. |
1da177e4c
|
964 965 |
*/ down_read(&old_mm->mmap_sem); |
999d9fc16
|
966 |
if (unlikely(old_mm->core_state)) { |
1da177e4c
|
967 968 969 970 971 972 973 974 975 |
up_read(&old_mm->mmap_sem); return -EINTR; } } task_lock(tsk); active_mm = tsk->active_mm; tsk->mm = mm; tsk->active_mm = mm; activate_mm(active_mm, mm); |
615d6e875
|
976 977 |
tsk->mm->vmacache_seqnum = 0; vmacache_flush(tsk); |
1da177e4c
|
978 |
task_unlock(tsk); |
1da177e4c
|
979 980 |
if (old_mm) { up_read(&old_mm->mmap_sem); |
7dddb12c6
|
981 |
BUG_ON(active_mm != old_mm); |
701085b21
|
982 |
setmax_mm_hiwater_rss(&tsk->signal->maxrss, old_mm); |
31a78f23b
|
983 |
mm_update_next_owner(old_mm); |
1da177e4c
|
984 985 986 987 988 989 990 991 992 993 994 995 996 |
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
|
997 |
static int de_thread(struct task_struct *tsk) |
1da177e4c
|
998 999 |
{ struct signal_struct *sig = tsk->signal; |
b2c903b87
|
1000 |
struct sighand_struct *oldsighand = tsk->sighand; |
1da177e4c
|
1001 |
spinlock_t *lock = &oldsighand->siglock; |
1da177e4c
|
1002 |
|
aafe6c2a2
|
1003 |
if (thread_group_empty(tsk)) |
1da177e4c
|
1004 1005 1006 1007 |
goto no_thread_group; /* * Kill all other threads in the thread group. |
1da177e4c
|
1008 |
*/ |
1da177e4c
|
1009 |
spin_lock_irq(lock); |
ed5d2cac1
|
1010 |
if (signal_group_exit(sig)) { |
1da177e4c
|
1011 1012 1013 1014 1015 |
/* * Another group action in progress, just * return so that the signal is processed. */ spin_unlock_irq(lock); |
1da177e4c
|
1016 1017 |
return -EAGAIN; } |
d344193a0
|
1018 |
|
ed5d2cac1
|
1019 |
sig->group_exit_task = tsk; |
d344193a0
|
1020 1021 1022 |
sig->notify_count = zap_other_threads(tsk); if (!thread_group_leader(tsk)) sig->notify_count--; |
1da177e4c
|
1023 |
|
d344193a0
|
1024 |
while (sig->notify_count) { |
d5bbd43d5
|
1025 |
__set_current_state(TASK_KILLABLE); |
1da177e4c
|
1026 1027 |
spin_unlock_irq(lock); schedule(); |
d5bbd43d5
|
1028 1029 |
if (unlikely(__fatal_signal_pending(tsk))) goto killed; |
1da177e4c
|
1030 1031 |
spin_lock_irq(lock); } |
1da177e4c
|
1032 1033 1034 1035 1036 1037 1038 |
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
|
1039 |
if (!thread_group_leader(tsk)) { |
8187926bd
|
1040 |
struct task_struct *leader = tsk->group_leader; |
6db840fa7
|
1041 |
|
6db840fa7
|
1042 |
for (;;) { |
e56fb2874
|
1043 |
threadgroup_change_begin(tsk); |
6db840fa7
|
1044 |
write_lock_irq(&tasklist_lock); |
dfcce791f
|
1045 1046 1047 1048 1049 |
/* * Do this under tasklist_lock to ensure that * exit_notify() can't miss ->group_exit_task */ sig->notify_count = -1; |
6db840fa7
|
1050 1051 |
if (likely(leader->exit_state)) break; |
d5bbd43d5
|
1052 |
__set_current_state(TASK_KILLABLE); |
6db840fa7
|
1053 |
write_unlock_irq(&tasklist_lock); |
e56fb2874
|
1054 |
threadgroup_change_end(tsk); |
6db840fa7
|
1055 |
schedule(); |
d5bbd43d5
|
1056 1057 |
if (unlikely(__fatal_signal_pending(tsk))) goto killed; |
6db840fa7
|
1058 |
} |
1da177e4c
|
1059 |
|
f5e902817
|
1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 |
/* * 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
|
1070 |
tsk->start_time = leader->start_time; |
266b7a021
|
1071 |
tsk->real_start_time = leader->real_start_time; |
f5e902817
|
1072 |
|
bac0abd61
|
1073 1074 |
BUG_ON(!same_thread_group(leader, tsk)); BUG_ON(has_group_leader_pid(tsk)); |
1da177e4c
|
1075 1076 1077 1078 1079 1080 |
/* * 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
|
1081 1082 |
/* Become a process group leader with the old leader's pid. |
c18258c6f
|
1083 1084 |
* The old leader becomes a thread of the this thread group. * Note: The old leader also uses this pid until release_task |
d73d65293
|
1085 1086 |
* is called. Odd but simple and correct. */ |
aafe6c2a2
|
1087 |
tsk->pid = leader->pid; |
3f4185483
|
1088 |
change_pid(tsk, PIDTYPE_PID, task_pid(leader)); |
aafe6c2a2
|
1089 1090 |
transfer_pid(leader, tsk, PIDTYPE_PGID); transfer_pid(leader, tsk, PIDTYPE_SID); |
9cd80bbb0
|
1091 |
|
aafe6c2a2
|
1092 |
list_replace_rcu(&leader->tasks, &tsk->tasks); |
9cd80bbb0
|
1093 |
list_replace_init(&leader->sibling, &tsk->sibling); |
1da177e4c
|
1094 |
|
aafe6c2a2
|
1095 1096 |
tsk->group_leader = tsk; leader->group_leader = tsk; |
de12a7878
|
1097 |
|
aafe6c2a2
|
1098 |
tsk->exit_signal = SIGCHLD; |
087806b12
|
1099 |
leader->exit_signal = -1; |
962b564cf
|
1100 1101 1102 |
BUG_ON(leader->exit_state != EXIT_ZOMBIE); leader->exit_state = EXIT_DEAD; |
eac1b5e57
|
1103 1104 1105 1106 1107 1108 1109 1110 |
/* * 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
|
1111 |
write_unlock_irq(&tasklist_lock); |
e56fb2874
|
1112 |
threadgroup_change_end(tsk); |
8187926bd
|
1113 1114 |
release_task(leader); |
ed5d2cac1
|
1115 |
} |
1da177e4c
|
1116 |
|
6db840fa7
|
1117 1118 |
sig->group_exit_task = NULL; sig->notify_count = 0; |
1da177e4c
|
1119 1120 |
no_thread_group: |
e63682534
|
1121 1122 |
/* we have changed execution domain */ tsk->exit_signal = SIGCHLD; |
1da177e4c
|
1123 |
exit_itimers(sig); |
cbaffba12
|
1124 |
flush_itimer_signals(); |
329f7dba5
|
1125 |
|
b2c903b87
|
1126 1127 |
if (atomic_read(&oldsighand->count) != 1) { struct sighand_struct *newsighand; |
1da177e4c
|
1128 |
/* |
b2c903b87
|
1129 1130 |
* This ->sighand is shared with the CLONE_SIGHAND * but not CLONE_THREAD task, switch to the new one. |
1da177e4c
|
1131 |
*/ |
b2c903b87
|
1132 1133 1134 |
newsighand = kmem_cache_alloc(sighand_cachep, GFP_KERNEL); if (!newsighand) return -ENOMEM; |
1da177e4c
|
1135 1136 1137 1138 1139 1140 |
atomic_set(&newsighand->count, 1); memcpy(newsighand->action, oldsighand->action, sizeof(newsighand->action)); write_lock_irq(&tasklist_lock); spin_lock(&oldsighand->siglock); |
aafe6c2a2
|
1141 |
rcu_assign_pointer(tsk->sighand, newsighand); |
1da177e4c
|
1142 1143 |
spin_unlock(&oldsighand->siglock); write_unlock_irq(&tasklist_lock); |
fba2afaae
|
1144 |
__cleanup_sighand(oldsighand); |
1da177e4c
|
1145 |
} |
aafe6c2a2
|
1146 |
BUG_ON(!thread_group_leader(tsk)); |
1da177e4c
|
1147 |
return 0; |
d5bbd43d5
|
1148 1149 1150 1151 1152 1153 1154 1155 |
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
|
1156 |
} |
0840a90d9
|
1157 |
|
59714d65d
|
1158 |
char *get_task_comm(char *buf, struct task_struct *tsk) |
1da177e4c
|
1159 1160 1161 1162 1163 |
{ /* buf must be at least sizeof(tsk->comm) in size */ task_lock(tsk); strncpy(buf, tsk->comm, sizeof(tsk->comm)); task_unlock(tsk); |
59714d65d
|
1164 |
return buf; |
1da177e4c
|
1165 |
} |
7d74f492e
|
1166 |
EXPORT_SYMBOL_GPL(get_task_comm); |
1da177e4c
|
1167 |
|
6a6d27de3
|
1168 1169 1170 1171 |
/* * These functions flushes out all traces of the currently running executable * so that a new one can be started */ |
82b897782
|
1172 |
void __set_task_comm(struct task_struct *tsk, const char *buf, bool exec) |
1da177e4c
|
1173 1174 |
{ task_lock(tsk); |
43d2b1132
|
1175 |
trace_task_rename(tsk, buf); |
1da177e4c
|
1176 1177 |
strlcpy(tsk->comm, buf, sizeof(tsk->comm)); task_unlock(tsk); |
82b897782
|
1178 |
perf_event_comm(tsk, exec); |
1da177e4c
|
1179 1180 1181 1182 |
} int flush_old_exec(struct linux_binprm * bprm) { |
221af7f87
|
1183 |
int retval; |
1da177e4c
|
1184 1185 1186 1187 1188 1189 1190 1191 |
/* * 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
|
1192 1193 1194 1195 1196 |
/* * Must be called _before_ exec_mmap() as bprm->mm is * not visibile until then. This also enables the update * to be lockless. */ |
925d1c401
|
1197 |
set_mm_exe_file(bprm->mm, bprm->file); |
6e399cd14
|
1198 |
|
1da177e4c
|
1199 |
/* |
1da177e4c
|
1200 1201 |
* Release all of the old mmap stuff */ |
3c77f8457
|
1202 |
acct_arg_size(bprm, 0); |
1da177e4c
|
1203 1204 |
retval = exec_mmap(bprm->mm); if (retval) |
fd8328be8
|
1205 |
goto out; |
1da177e4c
|
1206 1207 |
bprm->mm = NULL; /* We're using it now */ |
7ab02af42
|
1208 |
|
dac853ae8
|
1209 |
set_fs(USER_DS); |
b88fae644
|
1210 1211 |
current->flags &= ~(PF_RANDOMIZE | PF_FORKNOEXEC | PF_KTHREAD | PF_NOFREEZE | PF_NO_SETAFFINITY); |
7ab02af42
|
1212 1213 |
flush_thread(); current->personality &= ~bprm->per_clear; |
c1df5a637
|
1214 1215 1216 1217 1218 1219 1220 |
/* * 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
|
1221 1222 1223 1224 1225 1226 |
return 0; out: return retval; } EXPORT_SYMBOL(flush_old_exec); |
1b5d783c9
|
1227 1228 |
void would_dump(struct linux_binprm *bprm, struct file *file) { |
21245b863
|
1229 1230 1231 |
struct inode *inode = file_inode(file); if (inode_permission(inode, MAY_READ) < 0) { struct user_namespace *old, *user_ns; |
1b5d783c9
|
1232 |
bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP; |
21245b863
|
1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 |
/* 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
|
1245 1246 |
} EXPORT_SYMBOL(would_dump); |
221af7f87
|
1247 1248 |
void setup_new_exec(struct linux_binprm * bprm) { |
221af7f87
|
1249 |
arch_pick_mmap_layout(current->mm); |
1da177e4c
|
1250 1251 |
/* This is the point of no return */ |
1da177e4c
|
1252 |
current->sas_ss_sp = current->sas_ss_size = 0; |
8e96e3b7b
|
1253 |
if (uid_eq(current_euid(), current_uid()) && gid_eq(current_egid(), current_gid())) |
e579d2c25
|
1254 |
set_dumpable(current->mm, SUID_DUMP_USER); |
d6e711448
|
1255 |
else |
6c5d52382
|
1256 |
set_dumpable(current->mm, suid_dumpable); |
d6e711448
|
1257 |
|
e041e328c
|
1258 |
perf_event_exec(); |
82b897782
|
1259 |
__set_task_comm(current, kbasename(bprm->filename), true); |
1da177e4c
|
1260 |
|
0551fbd29
|
1261 1262 1263 1264 1265 |
/* Set the new mm task size. We have to do that late because it may * depend on TIF_32BIT which is only updated in flush_thread() on * some architectures like powerpc */ current->mm->task_size = TASK_SIZE; |
a6f76f23d
|
1266 |
/* install the new credentials */ |
8e96e3b7b
|
1267 1268 |
if (!uid_eq(bprm->cred->uid, current_euid()) || !gid_eq(bprm->cred->gid, current_egid())) { |
d2d56c5f5
|
1269 |
current->pdeath_signal = 0; |
1b5d783c9
|
1270 |
} else { |
1b5d783c9
|
1271 1272 |
if (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP) set_dumpable(current->mm, suid_dumpable); |
1da177e4c
|
1273 1274 1275 1276 |
} /* An exec changes our domain. We are no longer part of the thread group */ |
1da177e4c
|
1277 |
current->self_exec_id++; |
1da177e4c
|
1278 |
flush_signal_handlers(current, 0); |
1da177e4c
|
1279 |
} |
221af7f87
|
1280 |
EXPORT_SYMBOL(setup_new_exec); |
1da177e4c
|
1281 |
|
a6f76f23d
|
1282 |
/* |
a2a8474c3
|
1283 1284 1285 1286 1287 1288 1289 |
* Prepare credentials and lock ->cred_guard_mutex. * install_exec_creds() commits the new creds and drops the lock. * Or, if exec fails before, free_bprm() should release ->cred and * and unlock. */ int prepare_bprm_creds(struct linux_binprm *bprm) { |
9b1bf12d5
|
1290 |
if (mutex_lock_interruptible(¤t->signal->cred_guard_mutex)) |
a2a8474c3
|
1291 1292 1293 1294 1295 |
return -ERESTARTNOINTR; bprm->cred = prepare_exec_creds(); if (likely(bprm->cred)) return 0; |
9b1bf12d5
|
1296 |
mutex_unlock(¤t->signal->cred_guard_mutex); |
a2a8474c3
|
1297 1298 |
return -ENOMEM; } |
c4ad8f98b
|
1299 |
static void free_bprm(struct linux_binprm *bprm) |
a2a8474c3
|
1300 1301 1302 |
{ free_arg_pages(bprm); if (bprm->cred) { |
9b1bf12d5
|
1303 |
mutex_unlock(¤t->signal->cred_guard_mutex); |
a2a8474c3
|
1304 1305 |
abort_creds(bprm->cred); } |
63e46b95e
|
1306 1307 1308 1309 |
if (bprm->file) { allow_write_access(bprm->file); fput(bprm->file); } |
b66c59840
|
1310 1311 1312 |
/* If a binfmt changed the interp, free it. */ if (bprm->interp != bprm->filename) kfree(bprm->interp); |
a2a8474c3
|
1313 1314 |
kfree(bprm); } |
b66c59840
|
1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 |
int bprm_change_interp(char *interp, struct linux_binprm *bprm) { /* 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
|
1326 |
/* |
a6f76f23d
|
1327 1328 1329 1330 1331 1332 1333 1334 |
* 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
|
1335 1336 1337 1338 1339 1340 1341 1342 1343 |
/* * 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
|
1344 1345 |
/* * cred_guard_mutex must be held at least to this point to prevent |
a6f76f23d
|
1346 |
* ptrace_attach() from altering our determination of the task's |
a2a8474c3
|
1347 1348 |
* credentials; any time after this it may be unlocked. */ |
a6f76f23d
|
1349 |
security_bprm_committed_creds(bprm); |
9b1bf12d5
|
1350 |
mutex_unlock(¤t->signal->cred_guard_mutex); |
a6f76f23d
|
1351 1352 1353 1354 1355 |
} EXPORT_SYMBOL(install_exec_creds); /* * determine how safe it is to execute the proposed program |
9b1bf12d5
|
1356 |
* - the caller must hold ->cred_guard_mutex to protect against |
c2e1f2e30
|
1357 |
* PTRACE_ATTACH or seccomp thread-sync |
a6f76f23d
|
1358 |
*/ |
9e00cdb09
|
1359 |
static void check_unsafe_exec(struct linux_binprm *bprm) |
a6f76f23d
|
1360 |
{ |
0bf2f3aec
|
1361 |
struct task_struct *p = current, *t; |
f1191b50e
|
1362 |
unsigned n_fs; |
a6f76f23d
|
1363 |
|
4b9d33e6d
|
1364 |
if (p->ptrace) { |
e747b4ae3
|
1365 |
if (ptracer_capable(p, current_user_ns())) |
4b9d33e6d
|
1366 1367 1368 1369 |
bprm->unsafe |= LSM_UNSAFE_PTRACE_CAP; else bprm->unsafe |= LSM_UNSAFE_PTRACE; } |
a6f76f23d
|
1370 |
|
259e5e6c7
|
1371 1372 1373 1374 |
/* * This isn't strictly necessary, but it makes it harder for LSMs to * mess up. */ |
1d4457f99
|
1375 |
if (task_no_new_privs(current)) |
259e5e6c7
|
1376 |
bprm->unsafe |= LSM_UNSAFE_NO_NEW_PRIVS; |
83f62a2ea
|
1377 |
t = p; |
0bf2f3aec
|
1378 |
n_fs = 1; |
2a4419b5b
|
1379 |
spin_lock(&p->fs->lock); |
437f7fdb6
|
1380 |
rcu_read_lock(); |
83f62a2ea
|
1381 |
while_each_thread(p, t) { |
0bf2f3aec
|
1382 1383 |
if (t->fs == p->fs) n_fs++; |
0bf2f3aec
|
1384 |
} |
437f7fdb6
|
1385 |
rcu_read_unlock(); |
0bf2f3aec
|
1386 |
|
9e00cdb09
|
1387 |
if (p->fs->users > n_fs) |
a6f76f23d
|
1388 |
bprm->unsafe |= LSM_UNSAFE_SHARE; |
9e00cdb09
|
1389 1390 |
else p->fs->in_exec = 1; |
2a4419b5b
|
1391 |
spin_unlock(&p->fs->lock); |
a6f76f23d
|
1392 |
} |
8b01fc86b
|
1393 1394 1395 1396 1397 1398 |
static void bprm_fill_uid(struct linux_binprm *bprm) { struct inode *inode; unsigned int mode; kuid_t uid; kgid_t gid; |
cb6fd68fd
|
1399 1400 1401 1402 1403 1404 |
/* * 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
|
1405 1406 |
bprm->cred->euid = current_euid(); bprm->cred->egid = current_egid(); |
380cf5ba6
|
1407 |
if (!mnt_may_suid(bprm->file->f_path.mnt)) |
8b01fc86b
|
1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 |
return; if (task_no_new_privs(current)) return; inode = file_inode(bprm->file); mode = READ_ONCE(inode->i_mode); if (!(mode & (S_ISUID|S_ISGID))) return; /* Be careful if suid/sgid is set */ |
5955102c9
|
1419 |
inode_lock(inode); |
8b01fc86b
|
1420 1421 1422 1423 1424 |
/* reload atomically mode/uid/gid now that lock held */ mode = inode->i_mode; uid = inode->i_uid; gid = inode->i_gid; |
5955102c9
|
1425 |
inode_unlock(inode); |
8b01fc86b
|
1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 |
/* 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
|
1442 1443 |
/* * Fill the binprm structure from the inode. |
1da177e4c
|
1444 |
* Check permissions, then read the first 128 (BINPRM_BUF_SIZE) bytes |
a6f76f23d
|
1445 1446 |
* * This may be called multiple times for binary chains (scripts for example). |
1da177e4c
|
1447 1448 1449 |
*/ int prepare_binprm(struct linux_binprm *bprm) { |
1da177e4c
|
1450 |
int retval; |
8b01fc86b
|
1451 |
bprm_fill_uid(bprm); |
1da177e4c
|
1452 1453 |
/* fill in binprm security blob */ |
a6f76f23d
|
1454 |
retval = security_bprm_set_creds(bprm); |
1da177e4c
|
1455 1456 |
if (retval) return retval; |
a6f76f23d
|
1457 |
bprm->cred_prepared = 1; |
1da177e4c
|
1458 |
|
a6f76f23d
|
1459 1460 |
memset(bprm->buf, 0, BINPRM_BUF_SIZE); return kernel_read(bprm->file, 0, bprm->buf, BINPRM_BUF_SIZE); |
1da177e4c
|
1461 1462 1463 |
} EXPORT_SYMBOL(prepare_binprm); |
4fc75ff48
|
1464 1465 1466 1467 1468 |
/* * 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
|
1469 |
int remove_arg_zero(struct linux_binprm *bprm) |
1da177e4c
|
1470 |
{ |
b6a2fea39
|
1471 1472 1473 1474 |
int ret = 0; unsigned long offset; char *kaddr; struct page *page; |
4fc75ff48
|
1475 |
|
b6a2fea39
|
1476 1477 |
if (!bprm->argc) return 0; |
1da177e4c
|
1478 |
|
b6a2fea39
|
1479 1480 1481 1482 1483 1484 1485 |
do { offset = bprm->p & ~PAGE_MASK; page = get_arg_page(bprm, bprm->p, 0); if (!page) { ret = -EFAULT; goto out; } |
e8e3c3d66
|
1486 |
kaddr = kmap_atomic(page); |
4fc75ff48
|
1487 |
|
b6a2fea39
|
1488 1489 1490 |
for (; offset < PAGE_SIZE && kaddr[offset]; offset++, bprm->p++) ; |
4fc75ff48
|
1491 |
|
e8e3c3d66
|
1492 |
kunmap_atomic(kaddr); |
b6a2fea39
|
1493 |
put_arg_page(page); |
b6a2fea39
|
1494 |
} while (offset == PAGE_SIZE); |
4fc75ff48
|
1495 |
|
b6a2fea39
|
1496 1497 1498 |
bprm->p++; bprm->argc--; ret = 0; |
4fc75ff48
|
1499 |
|
b6a2fea39
|
1500 1501 |
out: return ret; |
1da177e4c
|
1502 |
} |
1da177e4c
|
1503 |
EXPORT_SYMBOL(remove_arg_zero); |
cb7b6b1cb
|
1504 1505 |
#define printable(c) (((c)=='\t') || ((c)==' ') || (0x20<=(c) && (c)<=0x7e)) |
1da177e4c
|
1506 1507 1508 |
/* * cycle the list of binary formats handler, until one recognizes the image */ |
3c456bfc4
|
1509 |
int search_binary_handler(struct linux_binprm *bprm) |
1da177e4c
|
1510 |
{ |
cb7b6b1cb
|
1511 |
bool need_retry = IS_ENABLED(CONFIG_MODULES); |
1da177e4c
|
1512 |
struct linux_binfmt *fmt; |
cb7b6b1cb
|
1513 |
int retval; |
1da177e4c
|
1514 |
|
d74026986
|
1515 |
/* This allows 4 levels of binfmt rewrites before failing hard. */ |
131b2f9f1
|
1516 |
if (bprm->recursion_depth > 5) |
d74026986
|
1517 |
return -ELOOP; |
1da177e4c
|
1518 1519 1520 |
retval = security_bprm_check(bprm); if (retval) return retval; |
1da177e4c
|
1521 |
retval = -ENOENT; |
cb7b6b1cb
|
1522 1523 1524 1525 1526 1527 1528 1529 |
retry: read_lock(&binfmt_lock); list_for_each_entry(fmt, &formats, lh) { if (!try_module_get(fmt->module)) continue; read_unlock(&binfmt_lock); bprm->recursion_depth++; retval = fmt->load_binary(bprm); |
19d860a14
|
1530 1531 |
read_lock(&binfmt_lock); put_binfmt(fmt); |
cb7b6b1cb
|
1532 |
bprm->recursion_depth--; |
19d860a14
|
1533 1534 1535 1536 1537 1538 1539 1540 |
if (retval < 0 && !bprm->mm) { /* we got to flush_old_exec() and failed after it */ read_unlock(&binfmt_lock); force_sigsegv(SIGSEGV, current); return retval; } if (retval != -ENOEXEC || !bprm->file) { read_unlock(&binfmt_lock); |
cb7b6b1cb
|
1541 |
return retval; |
1da177e4c
|
1542 |
} |
1da177e4c
|
1543 |
} |
cb7b6b1cb
|
1544 |
read_unlock(&binfmt_lock); |
19d860a14
|
1545 |
if (need_retry) { |
cb7b6b1cb
|
1546 1547 1548 |
if (printable(bprm->buf[0]) && printable(bprm->buf[1]) && printable(bprm->buf[2]) && printable(bprm->buf[3])) return retval; |
4e0621a07
|
1549 1550 |
if (request_module("binfmt-%04x", *(ushort *)(bprm->buf + 2)) < 0) return retval; |
cb7b6b1cb
|
1551 1552 1553 |
need_retry = false; goto retry; } |
1da177e4c
|
1554 1555 |
return retval; } |
1da177e4c
|
1556 |
EXPORT_SYMBOL(search_binary_handler); |
5d1baf3b6
|
1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 |
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
|
1570 |
audit_bprm(bprm); |
5d1baf3b6
|
1571 1572 |
trace_sched_process_exec(current, old_pid, bprm); ptrace_event(PTRACE_EVENT_EXEC, old_vpid); |
9beb266f2
|
1573 |
proc_exec_connector(current); |
5d1baf3b6
|
1574 1575 1576 1577 |
} return ret; } |
1da177e4c
|
1578 1579 1580 |
/* * sys_execve() executes a new program. */ |
51f39a1f0
|
1581 1582 1583 1584 |
static int do_execveat_common(int fd, struct filename *filename, struct user_arg_ptr argv, struct user_arg_ptr envp, int flags) |
1da177e4c
|
1585 |
{ |
51f39a1f0
|
1586 |
char *pathbuf = NULL; |
1da177e4c
|
1587 1588 |
struct linux_binprm *bprm; struct file *file; |
3b1253880
|
1589 |
struct files_struct *displaced; |
1da177e4c
|
1590 |
int retval; |
72fa59970
|
1591 |
|
c4ad8f98b
|
1592 1593 |
if (IS_ERR(filename)) return PTR_ERR(filename); |
72fa59970
|
1594 1595 1596 1597 1598 1599 1600 |
/* * 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
|
1601 |
atomic_read(¤t_user()->processes) > rlimit(RLIMIT_NPROC)) { |
72fa59970
|
1602 1603 1604 1605 1606 1607 1608 |
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
|
1609 |
|
3b1253880
|
1610 |
retval = unshare_files(&displaced); |
fd8328be8
|
1611 1612 |
if (retval) goto out_ret; |
1da177e4c
|
1613 |
retval = -ENOMEM; |
11b0b5abb
|
1614 |
bprm = kzalloc(sizeof(*bprm), GFP_KERNEL); |
1da177e4c
|
1615 |
if (!bprm) |
fd8328be8
|
1616 |
goto out_files; |
1da177e4c
|
1617 |
|
a2a8474c3
|
1618 1619 |
retval = prepare_bprm_creds(bprm); if (retval) |
a6f76f23d
|
1620 |
goto out_free; |
498052bba
|
1621 |
|
9e00cdb09
|
1622 |
check_unsafe_exec(bprm); |
a2a8474c3
|
1623 |
current->in_execve = 1; |
a6f76f23d
|
1624 |
|
51f39a1f0
|
1625 |
file = do_open_execat(fd, filename, flags); |
1da177e4c
|
1626 1627 |
retval = PTR_ERR(file); if (IS_ERR(file)) |
498052bba
|
1628 |
goto out_unmark; |
1da177e4c
|
1629 1630 |
sched_exec(); |
1da177e4c
|
1631 |
bprm->file = file; |
51f39a1f0
|
1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 |
if (fd == AT_FDCWD || filename->name[0] == '/') { bprm->filename = filename->name; } else { if (filename->name[0] == '\0') pathbuf = kasprintf(GFP_TEMPORARY, "/dev/fd/%d", fd); else pathbuf = kasprintf(GFP_TEMPORARY, "/dev/fd/%d/%s", 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
|
1654 |
|
b6a2fea39
|
1655 1656 |
retval = bprm_mm_init(bprm); if (retval) |
63e46b95e
|
1657 |
goto out_unmark; |
1da177e4c
|
1658 |
|
b6a2fea39
|
1659 |
bprm->argc = count(argv, MAX_ARG_STRINGS); |
1da177e4c
|
1660 |
if ((retval = bprm->argc) < 0) |
a6f76f23d
|
1661 |
goto out; |
1da177e4c
|
1662 |
|
b6a2fea39
|
1663 |
bprm->envc = count(envp, MAX_ARG_STRINGS); |
1da177e4c
|
1664 |
if ((retval = bprm->envc) < 0) |
1da177e4c
|
1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 |
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; |
21245b863
|
1683 |
would_dump(bprm, bprm->file); |
5d1baf3b6
|
1684 |
retval = exec_binprm(bprm); |
a6f76f23d
|
1685 1686 |
if (retval < 0) goto out; |
1da177e4c
|
1687 |
|
a6f76f23d
|
1688 |
/* execve succeeded */ |
498052bba
|
1689 |
current->fs->in_exec = 0; |
f9ce1f1cd
|
1690 |
current->in_execve = 0; |
a6f76f23d
|
1691 |
acct_update_integrals(current); |
82727018b
|
1692 |
task_numa_free(current); |
a6f76f23d
|
1693 |
free_bprm(bprm); |
51f39a1f0
|
1694 |
kfree(pathbuf); |
c4ad8f98b
|
1695 |
putname(filename); |
a6f76f23d
|
1696 1697 1698 |
if (displaced) put_files_struct(displaced); return retval; |
1da177e4c
|
1699 |
|
a6f76f23d
|
1700 |
out: |
3c77f8457
|
1701 1702 1703 1704 |
if (bprm->mm) { acct_arg_size(bprm, 0); mmput(bprm->mm); } |
1da177e4c
|
1705 |
|
498052bba
|
1706 |
out_unmark: |
9e00cdb09
|
1707 |
current->fs->in_exec = 0; |
f9ce1f1cd
|
1708 |
current->in_execve = 0; |
a6f76f23d
|
1709 1710 |
out_free: |
08a6fac1c
|
1711 |
free_bprm(bprm); |
51f39a1f0
|
1712 |
kfree(pathbuf); |
1da177e4c
|
1713 |
|
fd8328be8
|
1714 |
out_files: |
3b1253880
|
1715 1716 |
if (displaced) reset_files_struct(displaced); |
1da177e4c
|
1717 |
out_ret: |
c4ad8f98b
|
1718 |
putname(filename); |
1da177e4c
|
1719 1720 |
return retval; } |
c4ad8f98b
|
1721 |
int do_execve(struct filename *filename, |
ba2d01629
|
1722 |
const char __user *const __user *__argv, |
da3d4c5fa
|
1723 |
const char __user *const __user *__envp) |
ba2d01629
|
1724 |
{ |
0e028465d
|
1725 1726 |
struct user_arg_ptr argv = { .ptr.native = __argv }; struct user_arg_ptr envp = { .ptr.native = __envp }; |
51f39a1f0
|
1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 |
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
|
1739 1740 1741 |
} #ifdef CONFIG_COMPAT |
c4ad8f98b
|
1742 |
static int compat_do_execve(struct filename *filename, |
38b983b34
|
1743 |
const compat_uptr_t __user *__argv, |
d03d26e58
|
1744 |
const compat_uptr_t __user *__envp) |
0e028465d
|
1745 1746 1747 1748 1749 1750 1751 1752 1753 |
{ struct user_arg_ptr argv = { .is_compat = true, .ptr.compat = __argv, }; struct user_arg_ptr envp = { .is_compat = true, .ptr.compat = __envp, }; |
51f39a1f0
|
1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 |
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
|
1771 |
} |
0e028465d
|
1772 |
#endif |
ba2d01629
|
1773 |
|
964ee7df9
|
1774 |
void set_binfmt(struct linux_binfmt *new) |
1da177e4c
|
1775 |
{ |
801460d0c
|
1776 1777 1778 1779 |
struct mm_struct *mm = current->mm; if (mm->binfmt) module_put(mm->binfmt->module); |
1da177e4c
|
1780 |
|
801460d0c
|
1781 |
mm->binfmt = new; |
964ee7df9
|
1782 1783 |
if (new) __module_get(new->module); |
1da177e4c
|
1784 |
} |
1da177e4c
|
1785 |
EXPORT_SYMBOL(set_binfmt); |
6c5d52382
|
1786 |
/* |
7288e1187
|
1787 |
* set_dumpable stores three-value SUID_DUMP_* into mm->flags. |
6c5d52382
|
1788 1789 1790 |
*/ void set_dumpable(struct mm_struct *mm, int value) { |
abacd2fe3
|
1791 |
unsigned long old, new; |
7288e1187
|
1792 1793 |
if (WARN_ON((unsigned)value > SUID_DUMP_ROOT)) return; |
abacd2fe3
|
1794 1795 |
do { old = ACCESS_ONCE(mm->flags); |
7288e1187
|
1796 |
new = (old & ~MMF_DUMPABLE_MASK) | value; |
abacd2fe3
|
1797 |
} while (cmpxchg(&mm->flags, old, new) != old); |
6c5d52382
|
1798 |
} |
6c5d52382
|
1799 |
|
38b983b34
|
1800 1801 1802 1803 1804 |
SYSCALL_DEFINE3(execve, const char __user *, filename, const char __user *const __user *, argv, const char __user *const __user *, envp) { |
c4ad8f98b
|
1805 |
return do_execve(getname(filename), argv, envp); |
38b983b34
|
1806 |
} |
51f39a1f0
|
1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 |
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
|
1820 |
#ifdef CONFIG_COMPAT |
625b1d7e8
|
1821 1822 1823 |
COMPAT_SYSCALL_DEFINE3(execve, const char __user *, filename, const compat_uptr_t __user *, argv, const compat_uptr_t __user *, envp) |
38b983b34
|
1824 |
{ |
c4ad8f98b
|
1825 |
return compat_do_execve(getname(filename), argv, envp); |
38b983b34
|
1826 |
} |
51f39a1f0
|
1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 |
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
|
1840 |
#endif |