Blame view
kernel/exit.c
43.2 KB
1da177e4c
|
1 2 3 4 5 |
/* * linux/kernel/exit.c * * Copyright (C) 1991, 1992 Linus Torvalds */ |
1da177e4c
|
6 7 8 |
#include <linux/mm.h> #include <linux/slab.h> #include <linux/interrupt.h> |
1da177e4c
|
9 |
#include <linux/module.h> |
c59ede7b7
|
10 |
#include <linux/capability.h> |
1da177e4c
|
11 12 13 |
#include <linux/completion.h> #include <linux/personality.h> #include <linux/tty.h> |
da9cbc873
|
14 |
#include <linux/iocontext.h> |
1da177e4c
|
15 16 17 18 |
#include <linux/key.h> #include <linux/security.h> #include <linux/cpu.h> #include <linux/acct.h> |
8f0ab5147
|
19 |
#include <linux/tsacct_kern.h> |
1da177e4c
|
20 |
#include <linux/file.h> |
9f3acc314
|
21 |
#include <linux/fdtable.h> |
80d26af89
|
22 |
#include <linux/freezer.h> |
1da177e4c
|
23 |
#include <linux/binfmts.h> |
ab516013a
|
24 |
#include <linux/nsproxy.h> |
84d737866
|
25 |
#include <linux/pid_namespace.h> |
1da177e4c
|
26 27 28 29 |
#include <linux/ptrace.h> #include <linux/profile.h> #include <linux/mount.h> #include <linux/proc_fs.h> |
49d769d52
|
30 |
#include <linux/kthread.h> |
1da177e4c
|
31 |
#include <linux/mempolicy.h> |
c757249af
|
32 |
#include <linux/taskstats_kern.h> |
ca74e92b4
|
33 |
#include <linux/delayacct.h> |
b4f48b636
|
34 |
#include <linux/cgroup.h> |
1da177e4c
|
35 |
#include <linux/syscalls.h> |
7ed20e1ad
|
36 |
#include <linux/signal.h> |
6a14c5c9d
|
37 |
#include <linux/posix-timers.h> |
9f46080c4
|
38 |
#include <linux/cn_proc.h> |
de5097c2e
|
39 |
#include <linux/mutex.h> |
0771dfefc
|
40 |
#include <linux/futex.h> |
b92ce5589
|
41 |
#include <linux/pipe_fs_i.h> |
fa84cb935
|
42 |
#include <linux/audit.h> /* for audit_free() */ |
83cc5ed3c
|
43 |
#include <linux/resource.h> |
0d67a46df
|
44 |
#include <linux/blkdev.h> |
6eaeeaba3
|
45 |
#include <linux/task_io_accounting_ops.h> |
30199f5a4
|
46 |
#include <linux/tracehook.h> |
5ad4e53bd
|
47 |
#include <linux/fs_struct.h> |
d84f4f992
|
48 |
#include <linux/init_task.h> |
cdd6c482c
|
49 |
#include <linux/perf_event.h> |
ad8d75fff
|
50 |
#include <trace/events/sched.h> |
24f1e32c6
|
51 |
#include <linux/hw_breakpoint.h> |
3d5992d2a
|
52 |
#include <linux/oom.h> |
54848d73f
|
53 |
#include <linux/writeback.h> |
404015308
|
54 |
#include <linux/shm.h> |
5c9a8750a
|
55 |
#include <linux/kcov.h> |
1da177e4c
|
56 57 58 59 60 |
#include <asm/uaccess.h> #include <asm/unistd.h> #include <asm/pgtable.h> #include <asm/mmu_context.h> |
d40e48e02
|
61 |
static void __unhash_process(struct task_struct *p, bool group_dead) |
1da177e4c
|
62 63 |
{ nr_threads--; |
50d75f8da
|
64 |
detach_pid(p, PIDTYPE_PID); |
d40e48e02
|
65 |
if (group_dead) { |
1da177e4c
|
66 67 |
detach_pid(p, PIDTYPE_PGID); detach_pid(p, PIDTYPE_SID); |
c97d98931
|
68 |
|
5e85d4abe
|
69 |
list_del_rcu(&p->tasks); |
9cd80bbb0
|
70 |
list_del_init(&p->sibling); |
909ea9646
|
71 |
__this_cpu_dec(process_counts); |
1da177e4c
|
72 |
} |
47e65328a
|
73 |
list_del_rcu(&p->thread_group); |
0c740d0af
|
74 |
list_del_rcu(&p->thread_node); |
1da177e4c
|
75 |
} |
6a14c5c9d
|
76 77 78 79 80 81 |
/* * This function expects the tasklist_lock write-locked. */ static void __exit_signal(struct task_struct *tsk) { struct signal_struct *sig = tsk->signal; |
d40e48e02
|
82 |
bool group_dead = thread_group_leader(tsk); |
6a14c5c9d
|
83 |
struct sighand_struct *sighand; |
4ada856fb
|
84 |
struct tty_struct *uninitialized_var(tty); |
6fac4829c
|
85 |
cputime_t utime, stime; |
6a14c5c9d
|
86 |
|
d11c563dd
|
87 |
sighand = rcu_dereference_check(tsk->sighand, |
db1466b3e
|
88 |
lockdep_tasklist_lock_is_held()); |
6a14c5c9d
|
89 90 91 |
spin_lock(&sighand->siglock); posix_cpu_timers_exit(tsk); |
d40e48e02
|
92 |
if (group_dead) { |
6a14c5c9d
|
93 |
posix_cpu_timers_exit_group(tsk); |
4ada856fb
|
94 95 |
tty = sig->tty; sig->tty = NULL; |
4a5999429
|
96 |
} else { |
6a14c5c9d
|
97 |
/* |
e0a702171
|
98 99 100 101 102 103 104 105 |
* This can only happen if the caller is de_thread(). * FIXME: this is the temporary hack, we should teach * posix-cpu-timers to handle this case correctly. */ if (unlikely(has_group_leader_pid(tsk))) posix_cpu_timers_exit_group(tsk); /* |
6a14c5c9d
|
106 107 108 |
* If there is any task waiting for the group exit * then notify it: */ |
d344193a0
|
109 |
if (sig->notify_count > 0 && !--sig->notify_count) |
6a14c5c9d
|
110 |
wake_up_process(sig->group_exit_task); |
6db840fa7
|
111 |
|
6a14c5c9d
|
112 113 |
if (tsk == sig->curr_target) sig->curr_target = next_thread(tsk); |
6a14c5c9d
|
114 |
} |
90ed9cbe7
|
115 |
/* |
26e75b5c3
|
116 117 118 119 |
* Accumulate here the counters for all threads as they die. We could * skip the group leader because it is the last user of signal_struct, * but we want to avoid the race with thread_group_cputime() which can * see the empty ->thread_head list. |
90ed9cbe7
|
120 121 |
*/ task_cputime(tsk, &utime, &stime); |
e78c34967
|
122 |
write_seqlock(&sig->stats_lock); |
90ed9cbe7
|
123 124 125 126 127 128 129 130 131 132 133 |
sig->utime += utime; sig->stime += stime; sig->gtime += task_gtime(tsk); sig->min_flt += tsk->min_flt; sig->maj_flt += tsk->maj_flt; sig->nvcsw += tsk->nvcsw; sig->nivcsw += tsk->nivcsw; sig->inblock += task_io_get_inblock(tsk); sig->oublock += task_io_get_oublock(tsk); task_io_accounting_add(&sig->ioac, &tsk->ioac); sig->sum_sched_runtime += tsk->se.sum_exec_runtime; |
b3ac022cb
|
134 |
sig->nr_threads--; |
d40e48e02
|
135 |
__unhash_process(tsk, group_dead); |
e78c34967
|
136 |
write_sequnlock(&sig->stats_lock); |
5876700cd
|
137 |
|
da7978b03
|
138 139 140 141 142 |
/* * Do this under ->siglock, we can race with another thread * doing sigqueue_free() if we have SIGQUEUE_PREALLOC signals. */ flush_sigqueue(&tsk->pending); |
a7e5328a0
|
143 |
tsk->sighand = NULL; |
6a14c5c9d
|
144 |
spin_unlock(&sighand->siglock); |
6a14c5c9d
|
145 |
|
a7e5328a0
|
146 |
__cleanup_sighand(sighand); |
a0be55dee
|
147 |
clear_tsk_thread_flag(tsk, TIF_SIGPENDING); |
d40e48e02
|
148 |
if (group_dead) { |
6a14c5c9d
|
149 |
flush_sigqueue(&sig->shared_pending); |
4ada856fb
|
150 |
tty_kref_put(tty); |
6a14c5c9d
|
151 152 |
} } |
8c7904a00
|
153 154 |
static void delayed_put_task_struct(struct rcu_head *rhp) { |
0a16b6075
|
155 |
struct task_struct *tsk = container_of(rhp, struct task_struct, rcu); |
4e231c796
|
156 |
perf_event_delayed_put(tsk); |
0a16b6075
|
157 158 |
trace_sched_process_free(tsk); put_task_struct(tsk); |
8c7904a00
|
159 |
} |
f470021ad
|
160 |
|
a0be55dee
|
161 |
void release_task(struct task_struct *p) |
1da177e4c
|
162 |
{ |
36c8b5868
|
163 |
struct task_struct *leader; |
1da177e4c
|
164 |
int zap_leader; |
1f09f9749
|
165 |
repeat: |
c69e8d9c0
|
166 |
/* don't need to get the RCU readlock here - the process is dead and |
d11c563dd
|
167 168 |
* can't be modifying its own credentials. But shut RCU-lockdep up */ rcu_read_lock(); |
c69e8d9c0
|
169 |
atomic_dec(&__task_cred(p)->user->processes); |
d11c563dd
|
170 |
rcu_read_unlock(); |
c69e8d9c0
|
171 |
|
60347f671
|
172 |
proc_flush_task(p); |
0203026b5
|
173 |
|
1da177e4c
|
174 |
write_lock_irq(&tasklist_lock); |
a288eecce
|
175 |
ptrace_release_task(p); |
1da177e4c
|
176 |
__exit_signal(p); |
35f5cad8c
|
177 |
|
1da177e4c
|
178 179 180 181 182 183 184 |
/* * If we are the last non-leader member of the thread * group, and the leader is zombie, then notify the * group leader's parent process. (if it wants notification.) */ zap_leader = 0; leader = p->group_leader; |
a0be55dee
|
185 186 |
if (leader != p && thread_group_empty(leader) && leader->exit_state == EXIT_ZOMBIE) { |
1da177e4c
|
187 188 189 190 |
/* * If we were the last child thread and the leader has * exited already, and the leader's parent ignores SIGCHLD, * then we are the one who should release the leader. |
dae33574d
|
191 |
*/ |
867734737
|
192 |
zap_leader = do_notify_parent(leader, leader->exit_signal); |
dae33574d
|
193 194 |
if (zap_leader) leader->exit_state = EXIT_DEAD; |
1da177e4c
|
195 |
} |
1da177e4c
|
196 |
write_unlock_irq(&tasklist_lock); |
1da177e4c
|
197 |
release_thread(p); |
8c7904a00
|
198 |
call_rcu(&p->rcu, delayed_put_task_struct); |
1da177e4c
|
199 200 201 202 203 |
p = leader; if (unlikely(zap_leader)) goto repeat; } |
1da177e4c
|
204 |
/* |
150593bf8
|
205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 |
* Note that if this function returns a valid task_struct pointer (!NULL) * task->usage must remain >0 for the duration of the RCU critical section. */ struct task_struct *task_rcu_dereference(struct task_struct **ptask) { struct sighand_struct *sighand; struct task_struct *task; /* * We need to verify that release_task() was not called and thus * delayed_put_task_struct() can't run and drop the last reference * before rcu_read_unlock(). We check task->sighand != NULL, * but we can read the already freed and reused memory. */ retry: task = rcu_dereference(*ptask); if (!task) return NULL; probe_kernel_address(&task->sighand, sighand); /* * Pairs with atomic_dec_and_test() in put_task_struct(). If this task * was already freed we can not miss the preceding update of this * pointer. */ smp_rmb(); if (unlikely(task != READ_ONCE(*ptask))) goto retry; /* * We've re-checked that "task == *ptask", now we have two different * cases: * * 1. This is actually the same task/task_struct. In this case * sighand != NULL tells us it is still alive. * * 2. This is another task which got the same memory for task_struct. * We can't know this of course, and we can not trust * sighand != NULL. * * In this case we actually return a random value, but this is * correct. * * If we return NULL - we can pretend that we actually noticed that * *ptask was updated when the previous task has exited. Or pretend * that probe_slab_address(&sighand) reads NULL. * * If we return the new task (because sighand is not NULL for any * reason) - this is fine too. This (new) task can't go away before * another gp pass. * * And note: We could even eliminate the false positive if re-read * task->sighand once again to avoid the falsely NULL. But this case * is very unlikely so we don't care. */ if (!sighand) return NULL; return task; } struct task_struct *try_get_task_struct(struct task_struct **ptask) { struct task_struct *task; rcu_read_lock(); task = task_rcu_dereference(ptask); if (task) get_task_struct(task); rcu_read_unlock(); return task; } /* |
1da177e4c
|
281 282 283 284 285 286 287 |
* Determine if a process group is "orphaned", according to the POSIX * definition in 2.2.2.52. Orphaned process groups are not to be affected * by terminal-generated stop signals. Newly orphaned process groups are * to receive a SIGHUP and a SIGCONT. * * "I ask you, have you ever known what it is to be an orphan?" */ |
a0be55dee
|
288 289 |
static int will_become_orphaned_pgrp(struct pid *pgrp, struct task_struct *ignored_task) |
1da177e4c
|
290 291 |
{ struct task_struct *p; |
1da177e4c
|
292 |
|
0475ac084
|
293 |
do_each_pid_task(pgrp, PIDTYPE_PGID, p) { |
05e83df62
|
294 295 296 |
if ((p == ignored_task) || (p->exit_state && thread_group_empty(p)) || is_global_init(p->real_parent)) |
1da177e4c
|
297 |
continue; |
05e83df62
|
298 |
|
0475ac084
|
299 |
if (task_pgrp(p->real_parent) != pgrp && |
05e83df62
|
300 301 |
task_session(p->real_parent) == task_session(p)) return 0; |
0475ac084
|
302 |
} while_each_pid_task(pgrp, PIDTYPE_PGID, p); |
05e83df62
|
303 304 |
return 1; |
1da177e4c
|
305 |
} |
3e7cd6c41
|
306 |
int is_current_pgrp_orphaned(void) |
1da177e4c
|
307 308 309 310 |
{ int retval; read_lock(&tasklist_lock); |
3e7cd6c41
|
311 |
retval = will_become_orphaned_pgrp(task_pgrp(current), NULL); |
1da177e4c
|
312 313 314 315 |
read_unlock(&tasklist_lock); return retval; } |
961c4675c
|
316 |
static bool has_stopped_jobs(struct pid *pgrp) |
1da177e4c
|
317 |
{ |
1da177e4c
|
318 |
struct task_struct *p; |
0475ac084
|
319 |
do_each_pid_task(pgrp, PIDTYPE_PGID, p) { |
961c4675c
|
320 321 |
if (p->signal->flags & SIGNAL_STOP_STOPPED) return true; |
0475ac084
|
322 |
} while_each_pid_task(pgrp, PIDTYPE_PGID, p); |
961c4675c
|
323 324 |
return false; |
1da177e4c
|
325 |
} |
f49ee505b
|
326 327 328 329 330 331 332 333 334 335 336 337 |
/* * Check to see if any process groups have become orphaned as * a result of our exiting, and if they have any stopped jobs, * send them a SIGHUP and then a SIGCONT. (POSIX 3.2.2.2) */ static void kill_orphaned_pgrp(struct task_struct *tsk, struct task_struct *parent) { struct pid *pgrp = task_pgrp(tsk); struct task_struct *ignored_task = tsk; if (!parent) |
a0be55dee
|
338 339 340 |
/* exit: our father is in a different pgrp than * we are and we were the only connection outside. */ |
f49ee505b
|
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 |
parent = tsk->real_parent; else /* reparent: our child is in a different pgrp than * we are, and it was the only connection outside. */ ignored_task = NULL; if (task_pgrp(parent) != pgrp && task_session(parent) == task_session(tsk) && will_become_orphaned_pgrp(pgrp, ignored_task) && has_stopped_jobs(pgrp)) { __kill_pgrp_info(SIGHUP, SEND_SIG_PRIV, pgrp); __kill_pgrp_info(SIGCONT, SEND_SIG_PRIV, pgrp); } } |
f98bafa06
|
356 |
#ifdef CONFIG_MEMCG |
cf475ad28
|
357 |
/* |
733eda7ac
|
358 |
* A task is exiting. If it owned this mm, find a new owner for the mm. |
cf475ad28
|
359 |
*/ |
cf475ad28
|
360 361 362 363 364 |
void mm_update_next_owner(struct mm_struct *mm) { struct task_struct *c, *g, *p = current; retry: |
733eda7ac
|
365 366 367 368 369 |
/* * If the exiting or execing task is not the owner, it's * someone else's problem. */ if (mm->owner != p) |
cf475ad28
|
370 |
return; |
733eda7ac
|
371 372 373 374 375 376 377 378 379 |
/* * The current owner is exiting/execing and there are no other * candidates. Do not leave the mm pointing to a possibly * freed task structure. */ if (atomic_read(&mm->mm_users) <= 1) { mm->owner = NULL; return; } |
cf475ad28
|
380 381 382 383 384 385 386 387 388 389 390 391 392 |
read_lock(&tasklist_lock); /* * Search in the children */ list_for_each_entry(c, &p->children, sibling) { if (c->mm == mm) goto assign_new_owner; } /* * Search in the siblings */ |
dea33cfd9
|
393 |
list_for_each_entry(c, &p->real_parent->children, sibling) { |
cf475ad28
|
394 395 396 397 398 |
if (c->mm == mm) goto assign_new_owner; } /* |
f87fb599a
|
399 |
* Search through everything else, we should not get here often. |
cf475ad28
|
400 |
*/ |
39af1765f
|
401 402 403 404 405 406 407 408 409 |
for_each_process(g) { if (g->flags & PF_KTHREAD) continue; for_each_thread(g, c) { if (c->mm == mm) goto assign_new_owner; if (c->mm) break; } |
f87fb599a
|
410 |
} |
cf475ad28
|
411 |
read_unlock(&tasklist_lock); |
31a78f23b
|
412 413 414 |
/* * We found no owner yet mm_users > 1: this implies that we are * most likely racing with swapoff (try_to_unuse()) or /proc or |
e5991371e
|
415 |
* ptrace or page migration (get_task_mm()). Mark owner as NULL. |
31a78f23b
|
416 |
*/ |
31a78f23b
|
417 |
mm->owner = NULL; |
cf475ad28
|
418 419 420 421 422 423 424 425 426 427 |
return; assign_new_owner: BUG_ON(c == p); get_task_struct(c); /* * The task_lock protects c->mm from changing. * We always want mm->owner->mm == mm */ task_lock(c); |
e5991371e
|
428 429 430 431 432 |
/* * Delay read_unlock() till we have the task_lock() * to ensure that c does not slip away underneath us */ read_unlock(&tasklist_lock); |
cf475ad28
|
433 434 435 436 437 |
if (c->mm != mm) { task_unlock(c); put_task_struct(c); goto retry; } |
cf475ad28
|
438 439 440 441 |
mm->owner = c; task_unlock(c); put_task_struct(c); } |
f98bafa06
|
442 |
#endif /* CONFIG_MEMCG */ |
cf475ad28
|
443 |
|
1da177e4c
|
444 445 446 447 |
/* * Turn us into a lazy TLB process if we * aren't already.. */ |
a0be55dee
|
448 |
static void exit_mm(struct task_struct *tsk) |
1da177e4c
|
449 450 |
{ struct mm_struct *mm = tsk->mm; |
b564daf80
|
451 |
struct core_state *core_state; |
1da177e4c
|
452 |
|
48d212a2e
|
453 |
mm_release(tsk, mm); |
1da177e4c
|
454 455 |
if (!mm) return; |
4fe7efdbd
|
456 |
sync_mm_rss(mm); |
1da177e4c
|
457 458 |
/* * Serialize with any possible pending coredump. |
999d9fc16
|
459 |
* We must hold mmap_sem around checking core_state |
1da177e4c
|
460 |
* and clearing tsk->mm. The core-inducing thread |
999d9fc16
|
461 |
* will increment ->nr_threads for each thread in the |
1da177e4c
|
462 463 464 |
* group with ->mm != NULL. */ down_read(&mm->mmap_sem); |
b564daf80
|
465 466 467 |
core_state = mm->core_state; if (core_state) { struct core_thread self; |
a0be55dee
|
468 |
|
1da177e4c
|
469 |
up_read(&mm->mmap_sem); |
1da177e4c
|
470 |
|
b564daf80
|
471 472 473 474 475 476 477 478 |
self.task = tsk; self.next = xchg(&core_state->dumper.next, &self); /* * Implies mb(), the result of xchg() must be visible * to core_state->dumper. */ if (atomic_dec_and_test(&core_state->nr_threads)) complete(&core_state->startup); |
1da177e4c
|
479 |
|
a94e2d408
|
480 481 482 483 |
for (;;) { set_task_state(tsk, TASK_UNINTERRUPTIBLE); if (!self.task) /* see coredump_finish() */ break; |
80d26af89
|
484 |
freezable_schedule(); |
a94e2d408
|
485 486 |
} __set_task_state(tsk, TASK_RUNNING); |
1da177e4c
|
487 488 489 |
down_read(&mm->mmap_sem); } atomic_inc(&mm->mm_count); |
125e18745
|
490 |
BUG_ON(mm != tsk->active_mm); |
1da177e4c
|
491 492 493 494 495 496 |
/* more a memory barrier than a real lock */ task_lock(tsk); tsk->mm = NULL; up_read(&mm->mmap_sem); enter_lazy_tlb(mm, current); task_unlock(tsk); |
cf475ad28
|
497 |
mm_update_next_owner(mm); |
1da177e4c
|
498 |
mmput(mm); |
c32b3cbe0
|
499 |
if (test_thread_flag(TIF_MEMDIE)) |
38531201c
|
500 |
exit_oom_victim(); |
1da177e4c
|
501 |
} |
c9dc05bfd
|
502 503 504 505 506 507 508 509 510 511 |
static struct task_struct *find_alive_thread(struct task_struct *p) { struct task_struct *t; for_each_thread(p, t) { if (!(t->flags & PF_EXITING)) return t; } return NULL; } |
1109909c7
|
512 513 514 515 516 517 518 519 520 |
static struct task_struct *find_child_reaper(struct task_struct *father) __releases(&tasklist_lock) __acquires(&tasklist_lock) { struct pid_namespace *pid_ns = task_active_pid_ns(father); struct task_struct *reaper = pid_ns->child_reaper; if (likely(reaper != father)) return reaper; |
c9dc05bfd
|
521 522 |
reaper = find_alive_thread(father); if (reaper) { |
1109909c7
|
523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 |
pid_ns->child_reaper = reaper; return reaper; } write_unlock_irq(&tasklist_lock); if (unlikely(pid_ns == &init_pid_ns)) { panic("Attempted to kill init! exitcode=0x%08x ", father->signal->group_exit_code ?: father->exit_code); } zap_pid_ns_processes(pid_ns); write_lock_irq(&tasklist_lock); return father; } |
1da177e4c
|
538 |
/* |
ebec18a6d
|
539 540 541 542 543 |
* When we die, we re-parent all our children, and try to: * 1. give them to another thread in our thread group, if such a member exists * 2. give it to the first ancestor process which prctl'd itself as a * child_subreaper for its children (like a service manager) * 3. give it to the init process (PID 1) in our pid namespace |
1da177e4c
|
544 |
*/ |
1109909c7
|
545 546 |
static struct task_struct *find_new_reaper(struct task_struct *father, struct task_struct *child_reaper) |
1da177e4c
|
547 |
{ |
c9dc05bfd
|
548 |
struct task_struct *thread, *reaper; |
1da177e4c
|
549 |
|
c9dc05bfd
|
550 551 |
thread = find_alive_thread(father); if (thread) |
950bbabb5
|
552 |
return thread; |
1da177e4c
|
553 |
|
7d24e2df5
|
554 |
if (father->signal->has_child_subreaper) { |
ebec18a6d
|
555 |
/* |
175aed3f8
|
556 557 558 |
* Find the first ->is_child_subreaper ancestor in our pid_ns. * We start from father to ensure we can not look into another * namespace, this is safe because all its threads are dead. |
ebec18a6d
|
559 |
*/ |
7d24e2df5
|
560 |
for (reaper = father; |
1109909c7
|
561 |
!same_thread_group(reaper, child_reaper); |
ebec18a6d
|
562 |
reaper = reaper->real_parent) { |
175aed3f8
|
563 564 |
/* call_usermodehelper() descendants need this check */ if (reaper == &init_task) |
ebec18a6d
|
565 566 567 |
break; if (!reaper->signal->is_child_subreaper) continue; |
c9dc05bfd
|
568 569 570 |
thread = find_alive_thread(reaper); if (thread) return thread; |
ebec18a6d
|
571 |
} |
1da177e4c
|
572 |
} |
762a24bee
|
573 |
|
1109909c7
|
574 |
return child_reaper; |
950bbabb5
|
575 |
} |
5dfc80be7
|
576 577 578 |
/* * Any that need to be release_task'd are put on the @dead list. */ |
9cd80bbb0
|
579 |
static void reparent_leader(struct task_struct *father, struct task_struct *p, |
5dfc80be7
|
580 581 |
struct list_head *dead) { |
2831096e2
|
582 |
if (unlikely(p->exit_state == EXIT_DEAD)) |
5dfc80be7
|
583 |
return; |
abd50b39e
|
584 |
/* We don't want people slaying init. */ |
5dfc80be7
|
585 586 587 |
p->exit_signal = SIGCHLD; /* If it has exited notify the new parent about this child's death. */ |
d21142ece
|
588 |
if (!p->ptrace && |
5dfc80be7
|
589 |
p->exit_state == EXIT_ZOMBIE && thread_group_empty(p)) { |
867734737
|
590 |
if (do_notify_parent(p, p->exit_signal)) { |
5dfc80be7
|
591 |
p->exit_state = EXIT_DEAD; |
dc2fd4b00
|
592 |
list_add(&p->ptrace_entry, dead); |
5dfc80be7
|
593 594 595 596 597 |
} } kill_orphaned_pgrp(p, father); } |
482a3767e
|
598 599 600 601 602 603 604 605 606 607 |
/* * This does two things: * * A. Make init inherit all the child processes * B. Check to see if any process groups have become orphaned * as a result of our exiting, and if they have any stopped * jobs, send them a SIGHUP and then a SIGCONT. (POSIX 3.2.2.2) */ static void forget_original_parent(struct task_struct *father, struct list_head *dead) |
1da177e4c
|
608 |
{ |
482a3767e
|
609 |
struct task_struct *p, *t, *reaper; |
762a24bee
|
610 |
|
7c8bd2322
|
611 |
if (unlikely(!list_empty(&father->ptraced))) |
482a3767e
|
612 |
exit_ptrace(father, dead); |
f470021ad
|
613 |
|
7c8bd2322
|
614 |
/* Can drop and reacquire tasklist_lock */ |
1109909c7
|
615 |
reaper = find_child_reaper(father); |
ad9e206ae
|
616 |
if (list_empty(&father->children)) |
482a3767e
|
617 |
return; |
1109909c7
|
618 619 |
reaper = find_new_reaper(father, reaper); |
2831096e2
|
620 |
list_for_each_entry(p, &father->children, sibling) { |
57a059187
|
621 |
for_each_thread(p, t) { |
9cd80bbb0
|
622 |
t->real_parent = reaper; |
57a059187
|
623 624 |
BUG_ON((!t->ptrace) != (t->parent == father)); if (likely(!t->ptrace)) |
9cd80bbb0
|
625 |
t->parent = t->real_parent; |
9cd80bbb0
|
626 627 628 |
if (t->pdeath_signal) group_send_sig_info(t->pdeath_signal, SEND_SIG_NOINFO, t); |
57a059187
|
629 |
} |
2831096e2
|
630 631 632 633 634 |
/* * If this is a threaded reparent there is no need to * notify anyone anything has happened. */ if (!same_thread_group(reaper, father)) |
482a3767e
|
635 |
reparent_leader(father, p, dead); |
1da177e4c
|
636 |
} |
2831096e2
|
637 |
list_splice_tail_init(&father->children, &reaper->children); |
1da177e4c
|
638 639 640 641 642 643 |
} /* * Send signals to all our closest relatives so that they know * to properly mourn us.. */ |
821c7de71
|
644 |
static void exit_notify(struct task_struct *tsk, int group_dead) |
1da177e4c
|
645 |
{ |
53c8f9f19
|
646 |
bool autoreap; |
482a3767e
|
647 648 |
struct task_struct *p, *n; LIST_HEAD(dead); |
1da177e4c
|
649 |
|
762a24bee
|
650 |
write_lock_irq(&tasklist_lock); |
482a3767e
|
651 |
forget_original_parent(tsk, &dead); |
821c7de71
|
652 653 |
if (group_dead) kill_orphaned_pgrp(tsk->group_leader, NULL); |
1da177e4c
|
654 |
|
45cdf5cc0
|
655 656 657 658 659 660 661 662 663 664 665 666 |
if (unlikely(tsk->ptrace)) { int sig = thread_group_leader(tsk) && thread_group_empty(tsk) && !ptrace_reparented(tsk) ? tsk->exit_signal : SIGCHLD; autoreap = do_notify_parent(tsk, sig); } else if (thread_group_leader(tsk)) { autoreap = thread_group_empty(tsk) && do_notify_parent(tsk, tsk->exit_signal); } else { autoreap = true; } |
1da177e4c
|
667 |
|
53c8f9f19
|
668 |
tsk->exit_state = autoreap ? EXIT_DEAD : EXIT_ZOMBIE; |
6c66e7dba
|
669 670 |
if (tsk->exit_state == EXIT_DEAD) list_add(&tsk->ptrace_entry, &dead); |
1da177e4c
|
671 |
|
9c3391684
|
672 673 |
/* mt-exec, de_thread() is waiting for group leader */ if (unlikely(tsk->signal->notify_count < 0)) |
6db840fa7
|
674 |
wake_up_process(tsk->signal->group_exit_task); |
1da177e4c
|
675 |
write_unlock_irq(&tasklist_lock); |
482a3767e
|
676 677 678 679 |
list_for_each_entry_safe(p, n, &dead, ptrace_entry) { list_del_init(&p->ptrace_entry); release_task(p); } |
1da177e4c
|
680 |
} |
e18eecb8b
|
681 682 683 684 685 |
#ifdef CONFIG_DEBUG_STACK_USAGE static void check_stack_usage(void) { static DEFINE_SPINLOCK(low_water_lock); static int lowest_to_date = THREAD_SIZE; |
e18eecb8b
|
686 |
unsigned long free; |
7c9f8861e
|
687 |
free = stack_not_used(current); |
e18eecb8b
|
688 689 690 691 692 693 |
if (free >= lowest_to_date) return; spin_lock(&low_water_lock); if (free < lowest_to_date) { |
627393d44
|
694 695 |
pr_info("%s (%d) used greatest stack depth: %lu bytes left ", |
a0be55dee
|
696 |
current->comm, task_pid_nr(current), free); |
e18eecb8b
|
697 698 699 700 701 702 703 |
lowest_to_date = free; } spin_unlock(&low_water_lock); } #else static inline void check_stack_usage(void) {} #endif |
9af6528ee
|
704 |
void __noreturn do_exit(long code) |
1da177e4c
|
705 706 707 |
{ struct task_struct *tsk = current; int group_dead; |
3f95aa81d
|
708 |
TASKS_RCU(int tasks_rcu_i); |
1da177e4c
|
709 710 |
profile_task_exit(tsk); |
5c9a8750a
|
711 |
kcov_task_exit(tsk); |
1da177e4c
|
712 |
|
73c101011
|
713 |
WARN_ON(blk_needs_flush_plug(tsk)); |
22e2c507c
|
714 |
|
1da177e4c
|
715 716 717 718 |
if (unlikely(in_interrupt())) panic("Aiee, killing interrupt handler!"); if (unlikely(!tsk->pid)) panic("Attempted to kill the idle task!"); |
1da177e4c
|
719 |
|
33dd94ae1
|
720 721 722 723 724 725 726 727 |
/* * If do_exit is called because this processes oopsed, it's possible * that get_fs() was left as KERNEL_DS, so reset it to USER_DS before * continuing. Amongst other possible reasons, this is to prevent * mm_release()->clear_child_tid() from writing to a user-controlled * kernel address. */ set_fs(USER_DS); |
a288eecce
|
728 |
ptrace_event(PTRACE_EVENT_EXIT, code); |
1da177e4c
|
729 |
|
e0e817392
|
730 |
validate_creds_for_do_exit(tsk); |
df164db5f
|
731 732 733 734 735 |
/* * We're taking recursive faults here in do_exit. Safest is to just * leave this task alone and wait for reboot. */ if (unlikely(tsk->flags & PF_EXITING)) { |
a0be55dee
|
736 737 |
pr_alert("Fixing recursive fault but reboot is needed! "); |
778e9a9c3
|
738 739 740 741 742 743 744 745 746 747 |
/* * We can do this unlocked here. The futex code uses * this flag just to verify whether the pi state * cleanup has been done or not. In the worst case it * loops once more. We pretend that the cleanup was * done as there is no way to return. Either the * OWNER_DIED bit is set by now or we push the blocked * task into the wait for ever nirwana as well. */ tsk->flags |= PF_EXITPIDONE; |
df164db5f
|
748 749 750 |
set_current_state(TASK_UNINTERRUPTIBLE); schedule(); } |
d12619b5f
|
751 |
exit_signals(tsk); /* sets PF_EXITING */ |
778e9a9c3
|
752 |
/* |
be3e78449
|
753 754 |
* Ensure that all new tsk->pi_lock acquisitions must observe * PF_EXITING. Serializes against futex.c:attach_to_pi_owner(). |
778e9a9c3
|
755 |
*/ |
d2ee7198c
|
756 |
smp_mb(); |
be3e78449
|
757 758 759 760 |
/* * Ensure that we must observe the pi_state in exit_mm() -> * mm_release() -> exit_pi_state_list(). */ |
1d6154825
|
761 |
raw_spin_unlock_wait(&tsk->pi_lock); |
1da177e4c
|
762 |
|
1dc0fffc4
|
763 |
if (unlikely(in_atomic())) { |
a0be55dee
|
764 765 766 767 |
pr_info("note: %s[%d] exited with preempt_count %d ", current->comm, task_pid_nr(current), preempt_count()); |
1dc0fffc4
|
768 769 |
preempt_count_set(PREEMPT_ENABLED); } |
1da177e4c
|
770 |
|
48d212a2e
|
771 772 773 |
/* sync mm's RSS info before statistics gathering */ if (tsk->mm) sync_mm_rss(tsk->mm); |
51229b495
|
774 |
acct_update_integrals(tsk); |
1da177e4c
|
775 |
group_dead = atomic_dec_and_test(&tsk->signal->live); |
c30689516
|
776 |
if (group_dead) { |
778e9a9c3
|
777 |
hrtimer_cancel(&tsk->signal->real_timer); |
25f407f0b
|
778 |
exit_itimers(tsk->signal); |
1f10206cf
|
779 780 |
if (tsk->mm) setmax_mm_hiwater_rss(&tsk->signal->maxrss, tsk->mm); |
c30689516
|
781 |
} |
f6ec29a42
|
782 |
acct_collect(code, group_dead); |
522ed7767
|
783 784 |
if (group_dead) tty_audit_exit(); |
a4ff8dba7
|
785 |
audit_free(tsk); |
115085ea0
|
786 |
|
48d212a2e
|
787 |
tsk->exit_code = code; |
115085ea0
|
788 |
taskstats_exit(tsk, group_dead); |
c757249af
|
789 |
|
1da177e4c
|
790 |
exit_mm(tsk); |
0e4648141
|
791 |
if (group_dead) |
f6ec29a42
|
792 |
acct_process(); |
0a16b6075
|
793 |
trace_sched_process_exit(tsk); |
1da177e4c
|
794 |
exit_sem(tsk); |
b34a6b1da
|
795 |
exit_shm(tsk); |
1ec7f1ddb
|
796 797 |
exit_files(tsk); exit_fs(tsk); |
c39df5fa3
|
798 799 |
if (group_dead) disassociate_ctty(1); |
8aac62706
|
800 |
exit_task_namespaces(tsk); |
ed3e694d7
|
801 |
exit_task_work(tsk); |
e64646946
|
802 |
exit_thread(tsk); |
0b3fcf178
|
803 804 805 806 807 808 809 810 |
/* * Flush inherited counters to the parent - before the parent * gets woken up by child-exit notifications. * * because of cgroup mode, must be called before cgroup_exit() */ perf_event_exit_task(tsk); |
8e5bfa8c1
|
811 |
sched_autogroup_exit_task(tsk); |
1ec41830e
|
812 |
cgroup_exit(tsk); |
1da177e4c
|
813 |
|
33b2fb303
|
814 |
/* |
24f1e32c6
|
815 816 |
* FIXME: do that only when needed, using sched_exit tracepoint */ |
7c8df2863
|
817 |
flush_ptrace_hw_breakpoint(tsk); |
33b2fb303
|
818 |
|
49f5903b4
|
819 |
TASKS_RCU(preempt_disable()); |
3f95aa81d
|
820 |
TASKS_RCU(tasks_rcu_i = __srcu_read_lock(&tasks_rcu_exit_srcu)); |
49f5903b4
|
821 |
TASKS_RCU(preempt_enable()); |
821c7de71
|
822 |
exit_notify(tsk, group_dead); |
ef9823939
|
823 |
proc_exit_connector(tsk); |
c11600e4f
|
824 |
mpol_put_task_policy(tsk); |
42b2dd0a0
|
825 |
#ifdef CONFIG_FUTEX |
c87e2837b
|
826 827 |
if (unlikely(current->pi_state_cache)) kfree(current->pi_state_cache); |
42b2dd0a0
|
828 |
#endif |
c87e2837b
|
829 |
/* |
9a11b49a8
|
830 |
* Make sure we are holding no locks: |
de5097c2e
|
831 |
*/ |
1b1d2fb44
|
832 |
debug_check_no_locks_held(); |
778e9a9c3
|
833 834 835 836 837 838 |
/* * We can do this unlocked here. The futex code uses this flag * just to verify whether the pi state cleanup has been done * or not. In the worst case it loops once more. */ tsk->flags |= PF_EXITPIDONE; |
1da177e4c
|
839 |
|
afc847b7d
|
840 |
if (tsk->io_context) |
b69f22920
|
841 |
exit_io_context(tsk); |
afc847b7d
|
842 |
|
b92ce5589
|
843 |
if (tsk->splice_pipe) |
4b8a8f1e4
|
844 |
free_pipe_info(tsk->splice_pipe); |
b92ce5589
|
845 |
|
5640f7685
|
846 847 |
if (tsk->task_frag.page) put_page(tsk->task_frag.page); |
e0e817392
|
848 |
validate_creds_for_do_exit(tsk); |
4bcb8232c
|
849 |
check_stack_usage(); |
7407251a0
|
850 |
preempt_disable(); |
54848d73f
|
851 852 |
if (tsk->nr_dirtied) __this_cpu_add(dirty_throttle_leaks, tsk->nr_dirtied); |
f41d911f8
|
853 |
exit_rcu(); |
3f95aa81d
|
854 |
TASKS_RCU(__srcu_read_unlock(&tasks_rcu_exit_srcu, tasks_rcu_i)); |
b5740f4b2
|
855 |
|
9af6528ee
|
856 |
do_task_dead(); |
1da177e4c
|
857 |
} |
012914dad
|
858 |
EXPORT_SYMBOL_GPL(do_exit); |
9402c95f3
|
859 |
void complete_and_exit(struct completion *comp, long code) |
1da177e4c
|
860 861 862 |
{ if (comp) complete(comp); |
55a101f8f
|
863 |
|
1da177e4c
|
864 865 |
do_exit(code); } |
1da177e4c
|
866 |
EXPORT_SYMBOL(complete_and_exit); |
754fe8d29
|
867 |
SYSCALL_DEFINE1(exit, int, error_code) |
1da177e4c
|
868 869 870 |
{ do_exit((error_code&0xff)<<8); } |
1da177e4c
|
871 872 873 874 |
/* * Take down every thread in the group. This is called by fatal signals * as well as by sys_exit_group (below). */ |
9402c95f3
|
875 |
void |
1da177e4c
|
876 877 |
do_group_exit(int exit_code) { |
bfc4b0890
|
878 |
struct signal_struct *sig = current->signal; |
1da177e4c
|
879 |
BUG_ON(exit_code & 0x80); /* core dumps don't get here */ |
bfc4b0890
|
880 881 |
if (signal_group_exit(sig)) exit_code = sig->group_exit_code; |
1da177e4c
|
882 |
else if (!thread_group_empty(current)) { |
1da177e4c
|
883 |
struct sighand_struct *const sighand = current->sighand; |
a0be55dee
|
884 |
|
1da177e4c
|
885 |
spin_lock_irq(&sighand->siglock); |
ed5d2cac1
|
886 |
if (signal_group_exit(sig)) |
1da177e4c
|
887 888 889 |
/* Another thread got here before we took the lock. */ exit_code = sig->group_exit_code; else { |
1da177e4c
|
890 |
sig->group_exit_code = exit_code; |
ed5d2cac1
|
891 |
sig->flags = SIGNAL_GROUP_EXIT; |
1da177e4c
|
892 893 894 |
zap_other_threads(current); } spin_unlock_irq(&sighand->siglock); |
1da177e4c
|
895 896 897 898 899 900 901 902 903 904 905 |
} do_exit(exit_code); /* NOTREACHED */ } /* * this kills every thread in the thread group. Note that any externally * wait4()-ing process will get the correct exit code - even if this * thread is not the thread group leader. */ |
754fe8d29
|
906 |
SYSCALL_DEFINE1(exit_group, int, error_code) |
1da177e4c
|
907 908 |
{ do_group_exit((error_code & 0xff) << 8); |
2ed7c03ec
|
909 910 |
/* NOTREACHED */ return 0; |
1da177e4c
|
911 |
} |
9e8ae01d1
|
912 913 |
struct wait_opts { enum pid_type wo_type; |
9e8ae01d1
|
914 |
int wo_flags; |
e1eb1ebcc
|
915 |
struct pid *wo_pid; |
9e8ae01d1
|
916 917 918 919 |
struct siginfo __user *wo_info; int __user *wo_stat; struct rusage __user *wo_rusage; |
0b7570e77
|
920 |
wait_queue_t child_wait; |
9e8ae01d1
|
921 922 |
int notask_error; }; |
989264f46
|
923 924 |
static inline struct pid *task_pid_type(struct task_struct *task, enum pid_type type) |
161550d74
|
925 |
{ |
989264f46
|
926 927 928 |
if (type != PIDTYPE_PID) task = task->group_leader; return task->pids[type].pid; |
161550d74
|
929 |
} |
989264f46
|
930 |
static int eligible_pid(struct wait_opts *wo, struct task_struct *p) |
1da177e4c
|
931 |
{ |
5c01ba49e
|
932 933 934 |
return wo->wo_type == PIDTYPE_MAX || task_pid_type(p, wo->wo_type) == wo->wo_pid; } |
1da177e4c
|
935 |
|
bf959931d
|
936 937 |
static int eligible_child(struct wait_opts *wo, bool ptrace, struct task_struct *p) |
5c01ba49e
|
938 939 940 |
{ if (!eligible_pid(wo, p)) return 0; |
bf959931d
|
941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 |
/* * Wait for all children (clone and not) if __WALL is set or * if it is traced by us. */ if (ptrace || (wo->wo_flags & __WALL)) return 1; /* * Otherwise, wait for clone children *only* if __WCLONE is set; * otherwise, wait for non-clone children *only*. * * Note: a "clone" child here is one that reports to its parent * using a signal other than SIGCHLD, or a non-leader thread which * we can only see if it is traced by us. */ if ((p->exit_signal != SIGCHLD) ^ !!(wo->wo_flags & __WCLONE)) |
1da177e4c
|
958 |
return 0; |
1da177e4c
|
959 |
|
14dd0b814
|
960 |
return 1; |
1da177e4c
|
961 |
} |
9e8ae01d1
|
962 963 |
static int wait_noreap_copyout(struct wait_opts *wo, struct task_struct *p, pid_t pid, uid_t uid, int why, int status) |
1da177e4c
|
964 |
{ |
9e8ae01d1
|
965 966 967 |
struct siginfo __user *infop; int retval = wo->wo_rusage ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; |
36c8b5868
|
968 |
|
1da177e4c
|
969 |
put_task_struct(p); |
9e8ae01d1
|
970 |
infop = wo->wo_info; |
b6fe2d117
|
971 972 973 974 975 976 977 978 979 980 981 982 983 984 |
if (infop) { if (!retval) retval = put_user(SIGCHLD, &infop->si_signo); if (!retval) retval = put_user(0, &infop->si_errno); if (!retval) retval = put_user((short)why, &infop->si_code); if (!retval) retval = put_user(pid, &infop->si_pid); if (!retval) retval = put_user(uid, &infop->si_uid); if (!retval) retval = put_user(status, &infop->si_status); } |
1da177e4c
|
985 986 987 988 989 990 991 992 993 994 995 |
if (!retval) retval = pid; return retval; } /* * Handle sys_wait4 work for one task in state EXIT_ZOMBIE. We hold * read_lock(&tasklist_lock) on entry. If we return zero, we still hold * the lock and this task is uninteresting. If we return nonzero, we have * released the lock and the system call should return. */ |
9e8ae01d1
|
996 |
static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p) |
1da177e4c
|
997 |
{ |
f6507f83b
|
998 |
int state, retval, status; |
6c5f3e7b4
|
999 |
pid_t pid = task_pid_vnr(p); |
43e13cc10
|
1000 |
uid_t uid = from_kuid_munged(current_user_ns(), task_uid(p)); |
9e8ae01d1
|
1001 |
struct siginfo __user *infop; |
1da177e4c
|
1002 |
|
9e8ae01d1
|
1003 |
if (!likely(wo->wo_flags & WEXITED)) |
98abed020
|
1004 |
return 0; |
9e8ae01d1
|
1005 |
if (unlikely(wo->wo_flags & WNOWAIT)) { |
1da177e4c
|
1006 |
int exit_code = p->exit_code; |
f3abd4f95
|
1007 |
int why; |
1da177e4c
|
1008 |
|
1da177e4c
|
1009 1010 |
get_task_struct(p); read_unlock(&tasklist_lock); |
1029a2b52
|
1011 |
sched_annotate_sleep(); |
1da177e4c
|
1012 1013 1014 1015 1016 1017 1018 |
if ((exit_code & 0x7f) == 0) { why = CLD_EXITED; status = exit_code >> 8; } else { why = (exit_code & 0x80) ? CLD_DUMPED : CLD_KILLED; status = exit_code & 0x7f; } |
9e8ae01d1
|
1019 |
return wait_noreap_copyout(wo, p, pid, uid, why, status); |
1da177e4c
|
1020 |
} |
1da177e4c
|
1021 |
/* |
abd50b39e
|
1022 |
* Move the task's state to DEAD/TRACE, only one thread can do this. |
1da177e4c
|
1023 |
*/ |
f6507f83b
|
1024 1025 |
state = (ptrace_reparented(p) && thread_group_leader(p)) ? EXIT_TRACE : EXIT_DEAD; |
abd50b39e
|
1026 |
if (cmpxchg(&p->exit_state, EXIT_ZOMBIE, state) != EXIT_ZOMBIE) |
1da177e4c
|
1027 |
return 0; |
986094dfe
|
1028 1029 1030 1031 1032 |
/* * We own this thread, nobody else can reap it. */ read_unlock(&tasklist_lock); sched_annotate_sleep(); |
f6507f83b
|
1033 |
|
befca9677
|
1034 |
/* |
f6507f83b
|
1035 |
* Check thread_group_leader() to exclude the traced sub-threads. |
befca9677
|
1036 |
*/ |
f6507f83b
|
1037 |
if (state == EXIT_DEAD && thread_group_leader(p)) { |
f953ccd00
|
1038 1039 |
struct signal_struct *sig = p->signal; struct signal_struct *psig = current->signal; |
1f10206cf
|
1040 |
unsigned long maxrss; |
0cf55e1ec
|
1041 |
cputime_t tgutime, tgstime; |
3795e1616
|
1042 |
|
1da177e4c
|
1043 1044 1045 1046 1047 1048 1049 1050 |
/* * The resource counters for the group leader are in its * own task_struct. Those for dead threads in the group * are in its signal_struct, as are those for the child * processes it has previously reaped. All these * accumulate in the parent's signal_struct c* fields. * * We don't bother to take a lock here to protect these |
f953ccd00
|
1051 1052 1053 1054 1055 1056 1057 |
* p->signal fields because the whole thread group is dead * and nobody can change them. * * psig->stats_lock also protects us from our sub-theads * which can reap other children at the same time. Until * we change k_getrusage()-like users to rely on this lock * we have to take ->siglock as well. |
0cf55e1ec
|
1058 |
* |
a0be55dee
|
1059 1060 1061 |
* We use thread_group_cputime_adjusted() to get times for * the thread group, which consolidates times for all threads * in the group including the group leader. |
1da177e4c
|
1062 |
*/ |
e80d0a1ae
|
1063 |
thread_group_cputime_adjusted(p, &tgutime, &tgstime); |
f953ccd00
|
1064 |
spin_lock_irq(¤t->sighand->siglock); |
e78c34967
|
1065 |
write_seqlock(&psig->stats_lock); |
648616343
|
1066 1067 |
psig->cutime += tgutime + sig->cutime; psig->cstime += tgstime + sig->cstime; |
6fac4829c
|
1068 |
psig->cgtime += task_gtime(p) + sig->gtime + sig->cgtime; |
3795e1616
|
1069 1070 1071 1072 1073 1074 1075 1076 |
psig->cmin_flt += p->min_flt + sig->min_flt + sig->cmin_flt; psig->cmaj_flt += p->maj_flt + sig->maj_flt + sig->cmaj_flt; psig->cnvcsw += p->nvcsw + sig->nvcsw + sig->cnvcsw; psig->cnivcsw += p->nivcsw + sig->nivcsw + sig->cnivcsw; |
6eaeeaba3
|
1077 1078 1079 1080 1081 1082 |
psig->cinblock += task_io_get_inblock(p) + sig->inblock + sig->cinblock; psig->coublock += task_io_get_oublock(p) + sig->oublock + sig->coublock; |
1f10206cf
|
1083 1084 1085 |
maxrss = max(sig->maxrss, sig->cmaxrss); if (psig->cmaxrss < maxrss) psig->cmaxrss = maxrss; |
5995477ab
|
1086 1087 |
task_io_accounting_add(&psig->ioac, &p->ioac); task_io_accounting_add(&psig->ioac, &sig->ioac); |
e78c34967
|
1088 |
write_sequnlock(&psig->stats_lock); |
f953ccd00
|
1089 |
spin_unlock_irq(¤t->sighand->siglock); |
1da177e4c
|
1090 |
} |
9e8ae01d1
|
1091 1092 |
retval = wo->wo_rusage ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; |
1da177e4c
|
1093 1094 |
status = (p->signal->flags & SIGNAL_GROUP_EXIT) ? p->signal->group_exit_code : p->exit_code; |
9e8ae01d1
|
1095 1096 1097 1098 |
if (!retval && wo->wo_stat) retval = put_user(status, wo->wo_stat); infop = wo->wo_info; |
1da177e4c
|
1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 |
if (!retval && infop) retval = put_user(SIGCHLD, &infop->si_signo); if (!retval && infop) retval = put_user(0, &infop->si_errno); if (!retval && infop) { int why; if ((status & 0x7f) == 0) { why = CLD_EXITED; status >>= 8; } else { why = (status & 0x80) ? CLD_DUMPED : CLD_KILLED; status &= 0x7f; } retval = put_user((short)why, &infop->si_code); if (!retval) retval = put_user(status, &infop->si_status); } if (!retval && infop) |
3a515e4a6
|
1118 |
retval = put_user(pid, &infop->si_pid); |
1da177e4c
|
1119 |
if (!retval && infop) |
c69e8d9c0
|
1120 |
retval = put_user(uid, &infop->si_uid); |
2f4e6e2a8
|
1121 |
if (!retval) |
3a515e4a6
|
1122 |
retval = pid; |
2f4e6e2a8
|
1123 |
|
b43606905
|
1124 |
if (state == EXIT_TRACE) { |
1da177e4c
|
1125 |
write_lock_irq(&tasklist_lock); |
2f4e6e2a8
|
1126 1127 |
/* We dropped tasklist, ptracer could die and untrace */ ptrace_unlink(p); |
b43606905
|
1128 1129 1130 1131 1132 |
/* If parent wants a zombie, don't release it now */ state = EXIT_ZOMBIE; if (do_notify_parent(p, p->exit_signal)) state = EXIT_DEAD; |
abd50b39e
|
1133 |
p->exit_state = state; |
1da177e4c
|
1134 1135 |
write_unlock_irq(&tasklist_lock); } |
abd50b39e
|
1136 |
if (state == EXIT_DEAD) |
1da177e4c
|
1137 |
release_task(p); |
2f4e6e2a8
|
1138 |
|
1da177e4c
|
1139 1140 |
return retval; } |
90bc8d8b1
|
1141 1142 1143 |
static int *task_stopped_code(struct task_struct *p, bool ptrace) { if (ptrace) { |
570ac9337
|
1144 |
if (task_is_traced(p) && !(p->jobctl & JOBCTL_LISTENING)) |
90bc8d8b1
|
1145 1146 1147 1148 1149 1150 1151 |
return &p->exit_code; } else { if (p->signal->flags & SIGNAL_STOP_STOPPED) return &p->signal->group_exit_code; } return NULL; } |
19e274630
|
1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 |
/** * wait_task_stopped - Wait for %TASK_STOPPED or %TASK_TRACED * @wo: wait options * @ptrace: is the wait for ptrace * @p: task to wait for * * Handle sys_wait4() work for %p in state %TASK_STOPPED or %TASK_TRACED. * * CONTEXT: * read_lock(&tasklist_lock), which is released if return value is * non-zero. Also, grabs and releases @p->sighand->siglock. * * RETURNS: * 0 if wait condition didn't exist and search for other wait conditions * should continue. Non-zero return, -errno on failure and @p's pid on * success, implies that tasklist_lock is released and wait condition * search should terminate. |
1da177e4c
|
1169 |
*/ |
9e8ae01d1
|
1170 1171 |
static int wait_task_stopped(struct wait_opts *wo, int ptrace, struct task_struct *p) |
1da177e4c
|
1172 |
{ |
9e8ae01d1
|
1173 |
struct siginfo __user *infop; |
90bc8d8b1
|
1174 |
int retval, exit_code, *p_code, why; |
ee7c82da8
|
1175 |
uid_t uid = 0; /* unneeded, required by compiler */ |
c89507835
|
1176 |
pid_t pid; |
1da177e4c
|
1177 |
|
47918025e
|
1178 1179 1180 |
/* * Traditionally we see ptrace'd stopped tasks regardless of options. */ |
9e8ae01d1
|
1181 |
if (!ptrace && !(wo->wo_flags & WUNTRACED)) |
98abed020
|
1182 |
return 0; |
19e274630
|
1183 1184 |
if (!task_stopped_code(p, ptrace)) return 0; |
ee7c82da8
|
1185 1186 |
exit_code = 0; spin_lock_irq(&p->sighand->siglock); |
90bc8d8b1
|
1187 1188 |
p_code = task_stopped_code(p, ptrace); if (unlikely(!p_code)) |
ee7c82da8
|
1189 |
goto unlock_sig; |
90bc8d8b1
|
1190 |
exit_code = *p_code; |
ee7c82da8
|
1191 1192 |
if (!exit_code) goto unlock_sig; |
9e8ae01d1
|
1193 |
if (!unlikely(wo->wo_flags & WNOWAIT)) |
90bc8d8b1
|
1194 |
*p_code = 0; |
ee7c82da8
|
1195 |
|
8ca937a66
|
1196 |
uid = from_kuid_munged(current_user_ns(), task_uid(p)); |
ee7c82da8
|
1197 1198 1199 |
unlock_sig: spin_unlock_irq(&p->sighand->siglock); if (!exit_code) |
1da177e4c
|
1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 |
return 0; /* * Now we are pretty sure this task is interesting. * Make sure it doesn't get reaped out from under us while we * give up the lock and then examine it below. We don't want to * keep holding onto the tasklist_lock while we call getrusage and * possibly take page faults for user memory. */ get_task_struct(p); |
6c5f3e7b4
|
1210 |
pid = task_pid_vnr(p); |
f470021ad
|
1211 |
why = ptrace ? CLD_TRAPPED : CLD_STOPPED; |
1da177e4c
|
1212 |
read_unlock(&tasklist_lock); |
1029a2b52
|
1213 |
sched_annotate_sleep(); |
1da177e4c
|
1214 |
|
9e8ae01d1
|
1215 1216 1217 1218 1219 1220 1221 |
if (unlikely(wo->wo_flags & WNOWAIT)) return wait_noreap_copyout(wo, p, pid, uid, why, exit_code); retval = wo->wo_rusage ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; if (!retval && wo->wo_stat) retval = put_user((exit_code << 8) | 0x7f, wo->wo_stat); |
1da177e4c
|
1222 |
|
9e8ae01d1
|
1223 |
infop = wo->wo_info; |
1da177e4c
|
1224 1225 1226 1227 1228 |
if (!retval && infop) retval = put_user(SIGCHLD, &infop->si_signo); if (!retval && infop) retval = put_user(0, &infop->si_errno); if (!retval && infop) |
6efcae460
|
1229 |
retval = put_user((short)why, &infop->si_code); |
1da177e4c
|
1230 1231 1232 |
if (!retval && infop) retval = put_user(exit_code, &infop->si_status); if (!retval && infop) |
c89507835
|
1233 |
retval = put_user(pid, &infop->si_pid); |
1da177e4c
|
1234 |
if (!retval && infop) |
ee7c82da8
|
1235 |
retval = put_user(uid, &infop->si_uid); |
1da177e4c
|
1236 |
if (!retval) |
c89507835
|
1237 |
retval = pid; |
1da177e4c
|
1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 |
put_task_struct(p); BUG_ON(!retval); return retval; } /* * Handle do_wait work for one task in a live, non-stopped state. * read_lock(&tasklist_lock) on entry. If we return zero, we still hold * the lock and this task is uninteresting. If we return nonzero, we have * released the lock and the system call should return. */ |
9e8ae01d1
|
1250 |
static int wait_task_continued(struct wait_opts *wo, struct task_struct *p) |
1da177e4c
|
1251 1252 1253 1254 |
{ int retval; pid_t pid; uid_t uid; |
9e8ae01d1
|
1255 |
if (!unlikely(wo->wo_flags & WCONTINUED)) |
98abed020
|
1256 |
return 0; |
1da177e4c
|
1257 1258 1259 1260 1261 1262 1263 1264 1265 |
if (!(p->signal->flags & SIGNAL_STOP_CONTINUED)) return 0; spin_lock_irq(&p->sighand->siglock); /* Re-check with the lock held. */ if (!(p->signal->flags & SIGNAL_STOP_CONTINUED)) { spin_unlock_irq(&p->sighand->siglock); return 0; } |
9e8ae01d1
|
1266 |
if (!unlikely(wo->wo_flags & WNOWAIT)) |
1da177e4c
|
1267 |
p->signal->flags &= ~SIGNAL_STOP_CONTINUED; |
8ca937a66
|
1268 |
uid = from_kuid_munged(current_user_ns(), task_uid(p)); |
1da177e4c
|
1269 |
spin_unlock_irq(&p->sighand->siglock); |
6c5f3e7b4
|
1270 |
pid = task_pid_vnr(p); |
1da177e4c
|
1271 1272 |
get_task_struct(p); read_unlock(&tasklist_lock); |
1029a2b52
|
1273 |
sched_annotate_sleep(); |
1da177e4c
|
1274 |
|
9e8ae01d1
|
1275 1276 1277 |
if (!wo->wo_info) { retval = wo->wo_rusage ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; |
1da177e4c
|
1278 |
put_task_struct(p); |
9e8ae01d1
|
1279 1280 |
if (!retval && wo->wo_stat) retval = put_user(0xffff, wo->wo_stat); |
1da177e4c
|
1281 |
if (!retval) |
3a515e4a6
|
1282 |
retval = pid; |
1da177e4c
|
1283 |
} else { |
9e8ae01d1
|
1284 1285 |
retval = wait_noreap_copyout(wo, p, pid, uid, CLD_CONTINUED, SIGCONT); |
1da177e4c
|
1286 1287 1288 1289 1290 |
BUG_ON(retval == 0); } return retval; } |
98abed020
|
1291 1292 1293 |
/* * Consider @p for a wait by @parent. * |
9e8ae01d1
|
1294 |
* -ECHILD should be in ->notask_error before the first call. |
98abed020
|
1295 1296 |
* Returns nonzero for a final return, when we have unlocked tasklist_lock. * Returns zero if the search for a child should continue; |
9e8ae01d1
|
1297 |
* then ->notask_error is 0 if @p is an eligible child, |
14dd0b814
|
1298 |
* or another error from security_task_wait(), or still -ECHILD. |
98abed020
|
1299 |
*/ |
b6e763f07
|
1300 1301 |
static int wait_consider_task(struct wait_opts *wo, int ptrace, struct task_struct *p) |
98abed020
|
1302 |
{ |
3245d6aca
|
1303 1304 1305 1306 1307 1308 |
/* * We can race with wait_task_zombie() from another thread. * Ensure that EXIT_ZOMBIE -> EXIT_DEAD/EXIT_TRACE transition * can't confuse the checks below. */ int exit_state = ACCESS_ONCE(p->exit_state); |
b3ab03160
|
1309 |
int ret; |
3245d6aca
|
1310 |
if (unlikely(exit_state == EXIT_DEAD)) |
b3ab03160
|
1311 |
return 0; |
bf959931d
|
1312 |
ret = eligible_child(wo, ptrace, p); |
14dd0b814
|
1313 |
if (!ret) |
98abed020
|
1314 |
return ret; |
a2322e1d2
|
1315 |
ret = security_task_wait(p); |
14dd0b814
|
1316 1317 1318 1319 1320 1321 1322 1323 |
if (unlikely(ret < 0)) { /* * If we have not yet seen any eligible child, * then let this error code replace -ECHILD. * A permission error will give the user a clue * to look for security policy problems, rather * than for mysterious wait bugs. */ |
9e8ae01d1
|
1324 1325 |
if (wo->notask_error) wo->notask_error = ret; |
78a3d9d56
|
1326 |
return 0; |
14dd0b814
|
1327 |
} |
3245d6aca
|
1328 |
if (unlikely(exit_state == EXIT_TRACE)) { |
50b8d2574
|
1329 |
/* |
abd50b39e
|
1330 1331 |
* ptrace == 0 means we are the natural parent. In this case * we should clear notask_error, debugger will notify us. |
50b8d2574
|
1332 |
*/ |
abd50b39e
|
1333 |
if (likely(!ptrace)) |
50b8d2574
|
1334 |
wo->notask_error = 0; |
823b018e5
|
1335 |
return 0; |
50b8d2574
|
1336 |
} |
823b018e5
|
1337 |
|
377d75daf
|
1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 |
if (likely(!ptrace) && unlikely(p->ptrace)) { /* * If it is traced by its real parent's group, just pretend * the caller is ptrace_do_wait() and reap this child if it * is zombie. * * This also hides group stop state from real parent; otherwise * a single stop can be reported twice as group and ptrace stop. * If a ptracer wants to distinguish these two events for its * own children it should create a separate process which takes * the role of real parent. */ if (!ptrace_reparented(p)) ptrace = 1; } |
45cb24a1d
|
1353 |
/* slay zombie? */ |
3245d6aca
|
1354 |
if (exit_state == EXIT_ZOMBIE) { |
9b84cca25
|
1355 |
/* we don't reap group leaders with subthreads */ |
7c733eb3e
|
1356 1357 1358 1359 1360 1361 1362 1363 1364 |
if (!delay_group_leader(p)) { /* * A zombie ptracee is only visible to its ptracer. * Notification and reaping will be cascaded to the * real parent when the ptracer detaches. */ if (unlikely(ptrace) || likely(!p->ptrace)) return wait_task_zombie(wo, p); } |
98abed020
|
1365 |
|
f470021ad
|
1366 |
/* |
9b84cca25
|
1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 |
* Allow access to stopped/continued state via zombie by * falling through. Clearing of notask_error is complex. * * When !@ptrace: * * If WEXITED is set, notask_error should naturally be * cleared. If not, subset of WSTOPPED|WCONTINUED is set, * so, if there are live subthreads, there are events to * wait for. If all subthreads are dead, it's still safe * to clear - this function will be called again in finite * amount time once all the subthreads are released and * will then return without clearing. * * When @ptrace: * * Stopped state is per-task and thus can't change once the * target task dies. Only continued and exited can happen. * Clear notask_error if WCONTINUED | WEXITED. */ if (likely(!ptrace) || (wo->wo_flags & (WCONTINUED | WEXITED))) wo->notask_error = 0; } else { /* * @p is alive and it's gonna stop, continue or exit, so * there always is something to wait for. |
f470021ad
|
1392 |
*/ |
9e8ae01d1
|
1393 |
wo->notask_error = 0; |
f470021ad
|
1394 |
} |
98abed020
|
1395 |
/* |
45cb24a1d
|
1396 1397 |
* Wait for stopped. Depending on @ptrace, different stopped state * is used and the two don't interact with each other. |
98abed020
|
1398 |
*/ |
19e274630
|
1399 1400 1401 |
ret = wait_task_stopped(wo, ptrace, p); if (ret) return ret; |
98abed020
|
1402 1403 |
/* |
45cb24a1d
|
1404 1405 1406 |
* Wait for continued. There's only one continued state and the * ptracer can consume it which can confuse the real parent. Don't * use WCONTINUED from ptracer. You don't need or want it. |
98abed020
|
1407 |
*/ |
9e8ae01d1
|
1408 |
return wait_task_continued(wo, p); |
98abed020
|
1409 1410 1411 1412 1413 |
} /* * Do the work of do_wait() for one thread in the group, @tsk. * |
9e8ae01d1
|
1414 |
* -ECHILD should be in ->notask_error before the first call. |
98abed020
|
1415 1416 |
* Returns nonzero for a final return, when we have unlocked tasklist_lock. * Returns zero if the search for a child should continue; then |
9e8ae01d1
|
1417 |
* ->notask_error is 0 if there were any eligible children, |
14dd0b814
|
1418 |
* or another error from security_task_wait(), or still -ECHILD. |
98abed020
|
1419 |
*/ |
9e8ae01d1
|
1420 |
static int do_wait_thread(struct wait_opts *wo, struct task_struct *tsk) |
98abed020
|
1421 1422 1423 1424 |
{ struct task_struct *p; list_for_each_entry(p, &tsk->children, sibling) { |
9cd80bbb0
|
1425 |
int ret = wait_consider_task(wo, 0, p); |
a0be55dee
|
1426 |
|
9cd80bbb0
|
1427 1428 |
if (ret) return ret; |
98abed020
|
1429 1430 1431 1432 |
} return 0; } |
9e8ae01d1
|
1433 |
static int ptrace_do_wait(struct wait_opts *wo, struct task_struct *tsk) |
98abed020
|
1434 1435 |
{ struct task_struct *p; |
f470021ad
|
1436 |
list_for_each_entry(p, &tsk->ptraced, ptrace_entry) { |
b6e763f07
|
1437 |
int ret = wait_consider_task(wo, 1, p); |
a0be55dee
|
1438 |
|
f470021ad
|
1439 |
if (ret) |
98abed020
|
1440 |
return ret; |
98abed020
|
1441 1442 1443 1444 |
} return 0; } |
0b7570e77
|
1445 1446 1447 1448 1449 1450 |
static int child_wait_callback(wait_queue_t *wait, unsigned mode, int sync, void *key) { struct wait_opts *wo = container_of(wait, struct wait_opts, child_wait); struct task_struct *p = key; |
5c01ba49e
|
1451 |
if (!eligible_pid(wo, p)) |
0b7570e77
|
1452 |
return 0; |
b4fe51823
|
1453 1454 |
if ((wo->wo_flags & __WNOTHREAD) && wait->private != p->parent) return 0; |
0b7570e77
|
1455 1456 |
return default_wake_function(wait, mode, sync, key); } |
a7f0765ed
|
1457 1458 |
void __wake_up_parent(struct task_struct *p, struct task_struct *parent) { |
0b7570e77
|
1459 1460 |
__wake_up_sync_key(&parent->signal->wait_chldexit, TASK_INTERRUPTIBLE, 1, p); |
a7f0765ed
|
1461 |
} |
9e8ae01d1
|
1462 |
static long do_wait(struct wait_opts *wo) |
1da177e4c
|
1463 |
{ |
1da177e4c
|
1464 |
struct task_struct *tsk; |
98abed020
|
1465 |
int retval; |
1da177e4c
|
1466 |
|
9e8ae01d1
|
1467 |
trace_sched_process_wait(wo->wo_pid); |
0a16b6075
|
1468 |
|
0b7570e77
|
1469 1470 1471 |
init_waitqueue_func_entry(&wo->child_wait, child_wait_callback); wo->child_wait.private = current; add_wait_queue(¤t->signal->wait_chldexit, &wo->child_wait); |
1da177e4c
|
1472 |
repeat: |
98abed020
|
1473 |
/* |
3da56d166
|
1474 |
* If there is nothing that can match our criteria, just get out. |
9e8ae01d1
|
1475 1476 1477 |
* We will clear ->notask_error to zero if we see any child that * might later match our criteria, even if we are not able to reap * it yet. |
98abed020
|
1478 |
*/ |
64a16caf5
|
1479 |
wo->notask_error = -ECHILD; |
9e8ae01d1
|
1480 1481 |
if ((wo->wo_type < PIDTYPE_MAX) && (!wo->wo_pid || hlist_empty(&wo->wo_pid->tasks[wo->wo_type]))) |
64a16caf5
|
1482 |
goto notask; |
161550d74
|
1483 |
|
f95d39d10
|
1484 |
set_current_state(TASK_INTERRUPTIBLE); |
1da177e4c
|
1485 1486 1487 |
read_lock(&tasklist_lock); tsk = current; do { |
64a16caf5
|
1488 1489 1490 |
retval = do_wait_thread(wo, tsk); if (retval) goto end; |
9e8ae01d1
|
1491 |
|
64a16caf5
|
1492 1493 |
retval = ptrace_do_wait(wo, tsk); if (retval) |
98abed020
|
1494 |
goto end; |
98abed020
|
1495 |
|
9e8ae01d1
|
1496 |
if (wo->wo_flags & __WNOTHREAD) |
1da177e4c
|
1497 |
break; |
a3f6dfb72
|
1498 |
} while_each_thread(current, tsk); |
1da177e4c
|
1499 |
read_unlock(&tasklist_lock); |
f2cc3eb13
|
1500 |
|
64a16caf5
|
1501 |
notask: |
9e8ae01d1
|
1502 1503 |
retval = wo->notask_error; if (!retval && !(wo->wo_flags & WNOHANG)) { |
1da177e4c
|
1504 |
retval = -ERESTARTSYS; |
98abed020
|
1505 1506 1507 1508 |
if (!signal_pending(current)) { schedule(); goto repeat; } |
1da177e4c
|
1509 |
} |
1da177e4c
|
1510 |
end: |
f95d39d10
|
1511 |
__set_current_state(TASK_RUNNING); |
0b7570e77
|
1512 |
remove_wait_queue(¤t->signal->wait_chldexit, &wo->child_wait); |
1da177e4c
|
1513 1514 |
return retval; } |
17da2bd90
|
1515 1516 |
SYSCALL_DEFINE5(waitid, int, which, pid_t, upid, struct siginfo __user *, infop, int, options, struct rusage __user *, ru) |
1da177e4c
|
1517 |
{ |
9e8ae01d1
|
1518 |
struct wait_opts wo; |
161550d74
|
1519 1520 |
struct pid *pid = NULL; enum pid_type type; |
1da177e4c
|
1521 |
long ret; |
91c4e8ea8
|
1522 1523 |
if (options & ~(WNOHANG|WNOWAIT|WEXITED|WSTOPPED|WCONTINUED| __WNOTHREAD|__WCLONE|__WALL)) |
1da177e4c
|
1524 1525 1526 1527 1528 1529 |
return -EINVAL; if (!(options & (WEXITED|WSTOPPED|WCONTINUED))) return -EINVAL; switch (which) { case P_ALL: |
161550d74
|
1530 |
type = PIDTYPE_MAX; |
1da177e4c
|
1531 1532 |
break; case P_PID: |
161550d74
|
1533 1534 |
type = PIDTYPE_PID; if (upid <= 0) |
1da177e4c
|
1535 1536 1537 |
return -EINVAL; break; case P_PGID: |
161550d74
|
1538 1539 |
type = PIDTYPE_PGID; if (upid <= 0) |
1da177e4c
|
1540 |
return -EINVAL; |
1da177e4c
|
1541 1542 1543 1544 |
break; default: return -EINVAL; } |
161550d74
|
1545 1546 |
if (type < PIDTYPE_MAX) pid = find_get_pid(upid); |
9e8ae01d1
|
1547 1548 1549 1550 1551 1552 1553 1554 |
wo.wo_type = type; wo.wo_pid = pid; wo.wo_flags = options; wo.wo_info = infop; wo.wo_stat = NULL; wo.wo_rusage = ru; ret = do_wait(&wo); |
dfe16dfa4
|
1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 |
if (ret > 0) { ret = 0; } else if (infop) { /* * For a WNOHANG return, clear out all the fields * we would set so the user can easily tell the * difference. */ if (!ret) ret = put_user(0, &infop->si_signo); if (!ret) ret = put_user(0, &infop->si_errno); if (!ret) ret = put_user(0, &infop->si_code); if (!ret) ret = put_user(0, &infop->si_pid); if (!ret) ret = put_user(0, &infop->si_uid); if (!ret) ret = put_user(0, &infop->si_status); } |
161550d74
|
1577 |
put_pid(pid); |
1da177e4c
|
1578 1579 |
return ret; } |
754fe8d29
|
1580 1581 |
SYSCALL_DEFINE4(wait4, pid_t, upid, int __user *, stat_addr, int, options, struct rusage __user *, ru) |
1da177e4c
|
1582 |
{ |
9e8ae01d1
|
1583 |
struct wait_opts wo; |
161550d74
|
1584 1585 |
struct pid *pid = NULL; enum pid_type type; |
1da177e4c
|
1586 1587 1588 1589 1590 |
long ret; if (options & ~(WNOHANG|WUNTRACED|WCONTINUED| __WNOTHREAD|__WCLONE|__WALL)) return -EINVAL; |
161550d74
|
1591 1592 1593 1594 1595 1596 1597 1598 |
if (upid == -1) type = PIDTYPE_MAX; else if (upid < 0) { type = PIDTYPE_PGID; pid = find_get_pid(-upid); } else if (upid == 0) { type = PIDTYPE_PGID; |
2ae448efc
|
1599 |
pid = get_task_pid(current, PIDTYPE_PGID); |
161550d74
|
1600 1601 1602 1603 |
} else /* upid > 0 */ { type = PIDTYPE_PID; pid = find_get_pid(upid); } |
9e8ae01d1
|
1604 1605 1606 1607 1608 1609 1610 |
wo.wo_type = type; wo.wo_pid = pid; wo.wo_flags = options | WEXITED; wo.wo_info = NULL; wo.wo_stat = stat_addr; wo.wo_rusage = ru; ret = do_wait(&wo); |
161550d74
|
1611 |
put_pid(pid); |
1da177e4c
|
1612 |
|
1da177e4c
|
1613 1614 1615 1616 1617 1618 1619 1620 1621 |
return ret; } #ifdef __ARCH_WANT_SYS_WAITPID /* * sys_waitpid() remains for compatibility. waitpid() should be * implemented by calling sys_wait4() from libc.a. */ |
17da2bd90
|
1622 |
SYSCALL_DEFINE3(waitpid, pid_t, pid, int __user *, stat_addr, int, options) |
1da177e4c
|
1623 1624 1625 1626 1627 |
{ return sys_wait4(pid, stat_addr, options, NULL); } #endif |