Blame view
kernel/pid.c
14.9 KB
1da177e4c
|
1 2 3 |
/* * Generic pidhash and scalable, time-bounded PID allocator * |
6d49e352a
|
4 5 |
* (C) 2002-2003 Nadia Yvette Chambers, IBM * (C) 2004 Nadia Yvette Chambers, Oracle |
1da177e4c
|
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 |
* (C) 2002-2004 Ingo Molnar, Red Hat * * pid-structures are backing objects for tasks sharing a given ID to chain * against. There is very little to them aside from hashing them and * parking tasks using given ID's on a list. * * The hash is always changed with the tasklist_lock write-acquired, * and the hash is only accessed with the tasklist_lock at least * read-acquired, so there's no additional SMP locking needed here. * * We have a list of bitmap pages, which bitmaps represent the PID space. * Allocating and freeing PIDs is completely lockless. The worst-case * allocation scenario when all but one out of 1 million PIDs possible are * allocated already: the scanning of 32 list entries and at most PAGE_SIZE * bytes. The typical fastpath is a single successful setbit. Freeing is O(1). |
30e49c263
|
21 22 23 24 25 26 |
* * Pid namespaces: * (C) 2007 Pavel Emelyanov <xemul@openvz.org>, OpenVZ, SWsoft Inc. * (C) 2007 Sukadev Bhattiprolu <sukadev@us.ibm.com>, IBM * Many thanks to Oleg Nesterov for comments and help * |
1da177e4c
|
27 28 29 |
*/ #include <linux/mm.h> |
9984de1a5
|
30 |
#include <linux/export.h> |
1da177e4c
|
31 32 |
#include <linux/slab.h> #include <linux/init.h> |
82524746c
|
33 |
#include <linux/rculist.h> |
1da177e4c
|
34 35 |
#include <linux/bootmem.h> #include <linux/hash.h> |
61a58c6c2
|
36 |
#include <linux/pid_namespace.h> |
820e45db2
|
37 |
#include <linux/init_task.h> |
3eb07c8c8
|
38 |
#include <linux/syscalls.h> |
0bb80f240
|
39 |
#include <linux/proc_ns.h> |
0a01f2cc3
|
40 |
#include <linux/proc_fs.h> |
1da177e4c
|
41 |
|
8ef047aaa
|
42 43 |
#define pid_hashfn(nr, ns) \ hash_long((unsigned long)nr + (unsigned long)ns, pidhash_shift) |
92476d7fc
|
44 |
static struct hlist_head *pid_hash; |
2c85f51d2
|
45 |
static unsigned int pidhash_shift = 4; |
820e45db2
|
46 |
struct pid init_struct_pid = INIT_STRUCT_PID; |
1da177e4c
|
47 48 |
int pid_max = PID_MAX_DEFAULT; |
1da177e4c
|
49 50 51 52 53 |
#define RESERVED_PIDS 300 int pid_max_min = RESERVED_PIDS + 1; int pid_max_max = PID_MAX_LIMIT; |
61a58c6c2
|
54 55 |
static inline int mk_pid(struct pid_namespace *pid_ns, struct pidmap *map, int off) |
3fbc96486
|
56 |
{ |
61a58c6c2
|
57 |
return (map - pid_ns->pidmap)*BITS_PER_PAGE + off; |
3fbc96486
|
58 |
} |
1da177e4c
|
59 60 61 62 63 64 65 66 67 |
#define find_next_offset(map, off) \ find_next_zero_bit((map)->page, BITS_PER_PAGE, off) /* * PID-map pages start out as NULL, they get allocated upon * first use and are never deallocated. This way a low pid_max * value does not cause lots of bitmaps to be allocated, but * the scheme scales to up to 4 million PIDs, runtime. */ |
61a58c6c2
|
68 |
struct pid_namespace init_pid_ns = { |
9a575a92d
|
69 70 71 |
.kref = { .refcount = ATOMIC_INIT(2), }, |
3fbc96486
|
72 73 74 |
.pidmap = { [ 0 ... PIDMAP_ENTRIES-1] = { ATOMIC_INIT(BITS_PER_PAGE), NULL } }, |
84d737866
|
75 |
.last_pid = 0, |
8f75af44e
|
76 |
.nr_hashed = PIDNS_HASH_ADDING, |
faacbfd3a
|
77 78 |
.level = 0, .child_reaper = &init_task, |
49f4d8b93
|
79 |
.user_ns = &init_user_ns, |
435d5f4bb
|
80 |
.ns.inum = PROC_PID_INIT_INO, |
33c429405
|
81 82 83 |
#ifdef CONFIG_PID_NS .ns.ops = &pidns_operations, #endif |
3fbc96486
|
84 |
}; |
198fe21b0
|
85 |
EXPORT_SYMBOL_GPL(init_pid_ns); |
1da177e4c
|
86 |
|
92476d7fc
|
87 88 89 90 91 92 93 94 95 96 97 98 99 |
/* * Note: disable interrupts while the pidmap_lock is held as an * interrupt might come in and do read_lock(&tasklist_lock). * * If we don't disable interrupts there is a nasty deadlock between * detach_pid()->free_pid() and another cpu that does * spin_lock(&pidmap_lock) followed by an interrupt routine that does * read_lock(&tasklist_lock); * * After we clean up the tasklist_lock and know there are no * irq handlers that take it we can leave the interrupts enabled. * For now it is easier to be safe than to prove it can't happen. */ |
3fbc96486
|
100 |
|
1da177e4c
|
101 |
static __cacheline_aligned_in_smp DEFINE_SPINLOCK(pidmap_lock); |
b7127aa45
|
102 |
static void free_pidmap(struct upid *upid) |
1da177e4c
|
103 |
{ |
b7127aa45
|
104 105 106 |
int nr = upid->nr; struct pidmap *map = upid->ns->pidmap + nr / BITS_PER_PAGE; int offset = nr & BITS_PER_PAGE_MASK; |
1da177e4c
|
107 108 109 110 |
clear_bit(offset, map->page); atomic_inc(&map->nr_free); } |
5fdee8c4a
|
111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 |
/* * If we started walking pids at 'base', is 'a' seen before 'b'? */ static int pid_before(int base, int a, int b) { /* * This is the same as saying * * (a - base + MAXUINT) % MAXUINT < (b - base + MAXUINT) % MAXUINT * and that mapping orders 'a' and 'b' with respect to 'base'. */ return (unsigned)(a - base) < (unsigned)(b - base); } /* |
b8f566b04
|
126 127 128 |
* We might be racing with someone else trying to set pid_ns->last_pid * at the pid allocation time (there's also a sysctl for this, but racing * with this one is OK, see comment in kernel/pid_namespace.c about it). |
5fdee8c4a
|
129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 |
* We want the winner to have the "later" value, because if the * "earlier" value prevails, then a pid may get reused immediately. * * Since pids rollover, it is not sufficient to just pick the bigger * value. We have to consider where we started counting from. * * 'base' is the value of pid_ns->last_pid that we observed when * we started looking for a pid. * * 'pid' is the pid that we eventually found. */ static void set_last_pid(struct pid_namespace *pid_ns, int base, int pid) { int prev; int last_write = base; do { prev = last_write; last_write = cmpxchg(&pid_ns->last_pid, prev, pid); } while ((prev != last_write) && (pid_before(base, last_write, pid))); } |
61a58c6c2
|
149 |
static int alloc_pidmap(struct pid_namespace *pid_ns) |
1da177e4c
|
150 |
{ |
61a58c6c2
|
151 |
int i, offset, max_scan, pid, last = pid_ns->last_pid; |
6a1f3b845
|
152 |
struct pidmap *map; |
1da177e4c
|
153 154 155 156 157 |
pid = last + 1; if (pid >= pid_max) pid = RESERVED_PIDS; offset = pid & BITS_PER_PAGE_MASK; |
61a58c6c2
|
158 |
map = &pid_ns->pidmap[pid/BITS_PER_PAGE]; |
c52b0b91b
|
159 160 161 162 163 164 |
/* * If last_pid points into the middle of the map->page we * want to scan this bitmap block twice, the second time * we start with offset == 0 (or RESERVED_PIDS). */ max_scan = DIV_ROUND_UP(pid_max, BITS_PER_PAGE) - !offset; |
1da177e4c
|
165 166 |
for (i = 0; i <= max_scan; ++i) { if (unlikely(!map->page)) { |
3fbc96486
|
167 |
void *page = kzalloc(PAGE_SIZE, GFP_KERNEL); |
1da177e4c
|
168 169 170 171 |
/* * Free the page if someone raced with us * installing it: */ |
92476d7fc
|
172 |
spin_lock_irq(&pidmap_lock); |
7be6d991b
|
173 |
if (!map->page) { |
3fbc96486
|
174 |
map->page = page; |
7be6d991b
|
175 176 |
page = NULL; } |
92476d7fc
|
177 |
spin_unlock_irq(&pidmap_lock); |
7be6d991b
|
178 |
kfree(page); |
1da177e4c
|
179 180 181 182 |
if (unlikely(!map->page)) break; } if (likely(atomic_read(&map->nr_free))) { |
8db049b3d
|
183 |
for ( ; ; ) { |
1da177e4c
|
184 185 |
if (!test_and_set_bit(offset, map->page)) { atomic_dec(&map->nr_free); |
5fdee8c4a
|
186 |
set_last_pid(pid_ns, last, pid); |
1da177e4c
|
187 188 189 |
return pid; } offset = find_next_offset(map, offset); |
8db049b3d
|
190 191 |
if (offset >= BITS_PER_PAGE) break; |
61a58c6c2
|
192 |
pid = mk_pid(pid_ns, map, offset); |
8db049b3d
|
193 194 195 |
if (pid >= pid_max) break; } |
1da177e4c
|
196 |
} |
61a58c6c2
|
197 |
if (map < &pid_ns->pidmap[(pid_max-1)/BITS_PER_PAGE]) { |
1da177e4c
|
198 199 200 |
++map; offset = 0; } else { |
61a58c6c2
|
201 |
map = &pid_ns->pidmap[0]; |
1da177e4c
|
202 203 204 205 |
offset = RESERVED_PIDS; if (unlikely(last == offset)) break; } |
61a58c6c2
|
206 |
pid = mk_pid(pid_ns, map, offset); |
1da177e4c
|
207 208 209 |
} return -1; } |
c78193e9c
|
210 |
int next_pidmap(struct pid_namespace *pid_ns, unsigned int last) |
0804ef4b0
|
211 212 |
{ int offset; |
f40f50d3b
|
213 |
struct pidmap *map, *end; |
0804ef4b0
|
214 |
|
c78193e9c
|
215 216 |
if (last >= PID_MAX_LIMIT) return -1; |
0804ef4b0
|
217 |
offset = (last + 1) & BITS_PER_PAGE_MASK; |
61a58c6c2
|
218 219 |
map = &pid_ns->pidmap[(last + 1)/BITS_PER_PAGE]; end = &pid_ns->pidmap[PIDMAP_ENTRIES]; |
f40f50d3b
|
220 |
for (; map < end; map++, offset = 0) { |
0804ef4b0
|
221 222 223 224 |
if (unlikely(!map->page)) continue; offset = find_next_bit((map)->page, BITS_PER_PAGE, offset); if (offset < BITS_PER_PAGE) |
61a58c6c2
|
225 |
return mk_pid(pid_ns, map, offset); |
0804ef4b0
|
226 227 228 |
} return -1; } |
7ad5b3a50
|
229 |
void put_pid(struct pid *pid) |
92476d7fc
|
230 |
{ |
baf8f0f82
|
231 |
struct pid_namespace *ns; |
92476d7fc
|
232 233 |
if (!pid) return; |
baf8f0f82
|
234 |
|
8ef047aaa
|
235 |
ns = pid->numbers[pid->level].ns; |
92476d7fc
|
236 |
if ((atomic_read(&pid->count) == 1) || |
8ef047aaa
|
237 |
atomic_dec_and_test(&pid->count)) { |
baf8f0f82
|
238 |
kmem_cache_free(ns->pid_cachep, pid); |
b461cc038
|
239 |
put_pid_ns(ns); |
8ef047aaa
|
240 |
} |
92476d7fc
|
241 |
} |
bbf73147e
|
242 |
EXPORT_SYMBOL_GPL(put_pid); |
92476d7fc
|
243 244 245 246 247 248 |
static void delayed_put_pid(struct rcu_head *rhp) { struct pid *pid = container_of(rhp, struct pid, rcu); put_pid(pid); } |
7ad5b3a50
|
249 |
void free_pid(struct pid *pid) |
92476d7fc
|
250 251 |
{ /* We can be called with write_lock_irq(&tasklist_lock) held */ |
8ef047aaa
|
252 |
int i; |
92476d7fc
|
253 254 255 |
unsigned long flags; spin_lock_irqsave(&pidmap_lock, flags); |
0a01f2cc3
|
256 257 |
for (i = 0; i <= pid->level; i++) { struct upid *upid = pid->numbers + i; |
af4b8a83a
|
258 |
struct pid_namespace *ns = upid->ns; |
0a01f2cc3
|
259 |
hlist_del_rcu(&upid->pid_chain); |
af4b8a83a
|
260 |
switch(--ns->nr_hashed) { |
a60648851
|
261 |
case 2: |
af4b8a83a
|
262 263 264 265 266 267 268 |
case 1: /* When all that is left in the pid namespace * is the reaper wake up the reaper. The reaper * may be sleeping in zap_pid_ns_processes(). */ wake_up_process(ns->child_reaper); break; |
314a8ad0f
|
269 270 271 272 273 |
case PIDNS_HASH_ADDING: /* Handle a fork failure of the first process */ WARN_ON(ns->child_reaper); ns->nr_hashed = 0; /* fall through */ |
af4b8a83a
|
274 |
case 0: |
af4b8a83a
|
275 276 |
schedule_work(&ns->proc_work); break; |
5e1182deb
|
277 |
} |
0a01f2cc3
|
278 |
} |
92476d7fc
|
279 |
spin_unlock_irqrestore(&pidmap_lock, flags); |
8ef047aaa
|
280 |
for (i = 0; i <= pid->level; i++) |
b7127aa45
|
281 |
free_pidmap(pid->numbers + i); |
8ef047aaa
|
282 |
|
92476d7fc
|
283 284 |
call_rcu(&pid->rcu, delayed_put_pid); } |
8ef047aaa
|
285 |
struct pid *alloc_pid(struct pid_namespace *ns) |
92476d7fc
|
286 287 288 |
{ struct pid *pid; enum pid_type type; |
8ef047aaa
|
289 290 |
int i, nr; struct pid_namespace *tmp; |
198fe21b0
|
291 |
struct upid *upid; |
92476d7fc
|
292 |
|
baf8f0f82
|
293 |
pid = kmem_cache_alloc(ns->pid_cachep, GFP_KERNEL); |
92476d7fc
|
294 295 |
if (!pid) goto out; |
8ef047aaa
|
296 |
tmp = ns; |
0a01f2cc3
|
297 |
pid->level = ns->level; |
8ef047aaa
|
298 299 300 301 |
for (i = ns->level; i >= 0; i--) { nr = alloc_pidmap(tmp); if (nr < 0) goto out_free; |
92476d7fc
|
302 |
|
8ef047aaa
|
303 304 305 306 |
pid->numbers[i].nr = nr; pid->numbers[i].ns = tmp; tmp = tmp->parent; } |
0a01f2cc3
|
307 308 309 310 |
if (unlikely(is_child_reaper(pid))) { if (pid_ns_prepare_proc(ns)) goto out_free; } |
b461cc038
|
311 |
get_pid_ns(ns); |
92476d7fc
|
312 |
atomic_set(&pid->count, 1); |
92476d7fc
|
313 314 |
for (type = 0; type < PIDTYPE_MAX; ++type) INIT_HLIST_HEAD(&pid->tasks[type]); |
417e31524
|
315 |
upid = pid->numbers + ns->level; |
92476d7fc
|
316 |
spin_lock_irq(&pidmap_lock); |
c876ad768
|
317 |
if (!(ns->nr_hashed & PIDNS_HASH_ADDING)) |
5e1182deb
|
318 |
goto out_unlock; |
0a01f2cc3
|
319 |
for ( ; upid >= pid->numbers; --upid) { |
198fe21b0
|
320 321 |
hlist_add_head_rcu(&upid->pid_chain, &pid_hash[pid_hashfn(upid->nr, upid->ns)]); |
0a01f2cc3
|
322 323 |
upid->ns->nr_hashed++; } |
92476d7fc
|
324 325 326 327 |
spin_unlock_irq(&pidmap_lock); out: return pid; |
5e1182deb
|
328 |
out_unlock: |
6e6668845
|
329 |
spin_unlock_irq(&pidmap_lock); |
24c037ebf
|
330 |
put_pid_ns(ns); |
92476d7fc
|
331 |
out_free: |
b7127aa45
|
332 333 |
while (++i <= ns->level) free_pidmap(pid->numbers + i); |
8ef047aaa
|
334 |
|
baf8f0f82
|
335 |
kmem_cache_free(ns->pid_cachep, pid); |
92476d7fc
|
336 337 338 |
pid = NULL; goto out; } |
c876ad768
|
339 340 341 342 343 344 |
void disable_pid_allocation(struct pid_namespace *ns) { spin_lock_irq(&pidmap_lock); ns->nr_hashed &= ~PIDNS_HASH_ADDING; spin_unlock_irq(&pidmap_lock); } |
7ad5b3a50
|
345 |
struct pid *find_pid_ns(int nr, struct pid_namespace *ns) |
1da177e4c
|
346 |
{ |
198fe21b0
|
347 |
struct upid *pnr; |
b67bfe0d4
|
348 |
hlist_for_each_entry_rcu(pnr, |
198fe21b0
|
349 350 351 352 |
&pid_hash[pid_hashfn(nr, ns)], pid_chain) if (pnr->nr == nr && pnr->ns == ns) return container_of(pnr, struct pid, numbers[ns->level]); |
1da177e4c
|
353 |
|
1da177e4c
|
354 355 |
return NULL; } |
198fe21b0
|
356 |
EXPORT_SYMBOL_GPL(find_pid_ns); |
1da177e4c
|
357 |
|
8990571eb
|
358 359 |
struct pid *find_vpid(int nr) { |
17cf22c33
|
360 |
return find_pid_ns(nr, task_active_pid_ns(current)); |
8990571eb
|
361 362 |
} EXPORT_SYMBOL_GPL(find_vpid); |
e713d0dab
|
363 364 365 |
/* * attach_pid() must be called with the tasklist_lock write-held. */ |
819077398
|
366 |
void attach_pid(struct task_struct *task, enum pid_type type) |
1da177e4c
|
367 |
{ |
819077398
|
368 369 |
struct pid_link *link = &task->pids[type]; hlist_add_head_rcu(&link->node, &link->pid->tasks[type]); |
1da177e4c
|
370 |
} |
24336eaee
|
371 372 |
static void __change_pid(struct task_struct *task, enum pid_type type, struct pid *new) |
1da177e4c
|
373 |
{ |
92476d7fc
|
374 375 376 |
struct pid_link *link; struct pid *pid; int tmp; |
1da177e4c
|
377 |
|
92476d7fc
|
378 379 |
link = &task->pids[type]; pid = link->pid; |
1da177e4c
|
380 |
|
92476d7fc
|
381 |
hlist_del_rcu(&link->node); |
24336eaee
|
382 |
link->pid = new; |
1da177e4c
|
383 |
|
92476d7fc
|
384 385 386 |
for (tmp = PIDTYPE_MAX; --tmp >= 0; ) if (!hlist_empty(&pid->tasks[tmp])) return; |
1da177e4c
|
387 |
|
92476d7fc
|
388 |
free_pid(pid); |
1da177e4c
|
389 |
} |
24336eaee
|
390 391 392 393 394 395 396 397 398 |
void detach_pid(struct task_struct *task, enum pid_type type) { __change_pid(task, type, NULL); } void change_pid(struct task_struct *task, enum pid_type type, struct pid *pid) { __change_pid(task, type, pid); |
819077398
|
399 |
attach_pid(task, type); |
24336eaee
|
400 |
} |
c18258c6f
|
401 |
/* transfer_pid is an optimization of attach_pid(new), detach_pid(old) */ |
7ad5b3a50
|
402 |
void transfer_pid(struct task_struct *old, struct task_struct *new, |
c18258c6f
|
403 404 405 406 |
enum pid_type type) { new->pids[type].pid = old->pids[type].pid; hlist_replace_rcu(&old->pids[type].node, &new->pids[type].node); |
c18258c6f
|
407 |
} |
7ad5b3a50
|
408 |
struct task_struct *pid_task(struct pid *pid, enum pid_type type) |
1da177e4c
|
409 |
{ |
92476d7fc
|
410 411 412 |
struct task_struct *result = NULL; if (pid) { struct hlist_node *first; |
67bdbffd6
|
413 |
first = rcu_dereference_check(hlist_first_rcu(&pid->tasks[type]), |
db1466b3e
|
414 |
lockdep_tasklist_lock_is_held()); |
92476d7fc
|
415 416 417 418 419 |
if (first) result = hlist_entry(first, struct task_struct, pids[(type)].node); } return result; } |
eccba0689
|
420 |
EXPORT_SYMBOL(pid_task); |
1da177e4c
|
421 |
|
92476d7fc
|
422 |
/* |
9728e5d6e
|
423 |
* Must be called under rcu_read_lock(). |
92476d7fc
|
424 |
*/ |
17f98dcf6
|
425 |
struct task_struct *find_task_by_pid_ns(pid_t nr, struct pid_namespace *ns) |
92476d7fc
|
426 |
{ |
b3fbab057
|
427 428 429 |
rcu_lockdep_assert(rcu_read_lock_held(), "find_task_by_pid_ns() needs rcu_read_lock()" " protection"); |
17f98dcf6
|
430 |
return pid_task(find_pid_ns(nr, ns), PIDTYPE_PID); |
92476d7fc
|
431 |
} |
1da177e4c
|
432 |
|
228ebcbe6
|
433 434 |
struct task_struct *find_task_by_vpid(pid_t vnr) { |
17cf22c33
|
435 |
return find_task_by_pid_ns(vnr, task_active_pid_ns(current)); |
228ebcbe6
|
436 |
} |
228ebcbe6
|
437 |
|
1a657f78d
|
438 439 440 441 |
struct pid *get_task_pid(struct task_struct *task, enum pid_type type) { struct pid *pid; rcu_read_lock(); |
2ae448efc
|
442 443 |
if (type != PIDTYPE_PID) task = task->group_leader; |
1a657f78d
|
444 445 446 447 |
pid = get_pid(task->pids[type].pid); rcu_read_unlock(); return pid; } |
77c100c83
|
448 |
EXPORT_SYMBOL_GPL(get_task_pid); |
1a657f78d
|
449 |
|
7ad5b3a50
|
450 |
struct task_struct *get_pid_task(struct pid *pid, enum pid_type type) |
92476d7fc
|
451 452 453 454 455 456 457 458 |
{ struct task_struct *result; rcu_read_lock(); result = pid_task(pid, type); if (result) get_task_struct(result); rcu_read_unlock(); return result; |
1da177e4c
|
459 |
} |
77c100c83
|
460 |
EXPORT_SYMBOL_GPL(get_pid_task); |
1da177e4c
|
461 |
|
92476d7fc
|
462 |
struct pid *find_get_pid(pid_t nr) |
1da177e4c
|
463 464 |
{ struct pid *pid; |
92476d7fc
|
465 |
rcu_read_lock(); |
198fe21b0
|
466 |
pid = get_pid(find_vpid(nr)); |
92476d7fc
|
467 |
rcu_read_unlock(); |
1da177e4c
|
468 |
|
92476d7fc
|
469 |
return pid; |
1da177e4c
|
470 |
} |
339caf2a2
|
471 |
EXPORT_SYMBOL_GPL(find_get_pid); |
1da177e4c
|
472 |
|
7af572947
|
473 474 475 476 477 478 479 480 481 482 483 484 |
pid_t pid_nr_ns(struct pid *pid, struct pid_namespace *ns) { struct upid *upid; pid_t nr = 0; if (pid && ns->level <= pid->level) { upid = &pid->numbers[ns->level]; if (upid->ns == ns) nr = upid->nr; } return nr; } |
4f82f4573
|
485 |
EXPORT_SYMBOL_GPL(pid_nr_ns); |
7af572947
|
486 |
|
44c4e1b25
|
487 488 |
pid_t pid_vnr(struct pid *pid) { |
17cf22c33
|
489 |
return pid_nr_ns(pid, task_active_pid_ns(current)); |
44c4e1b25
|
490 491 |
} EXPORT_SYMBOL_GPL(pid_vnr); |
52ee2dfdd
|
492 493 |
pid_t __task_pid_nr_ns(struct task_struct *task, enum pid_type type, struct pid_namespace *ns) |
2f2a3a46f
|
494 |
{ |
52ee2dfdd
|
495 496 497 498 |
pid_t nr = 0; rcu_read_lock(); if (!ns) |
17cf22c33
|
499 |
ns = task_active_pid_ns(current); |
52ee2dfdd
|
500 501 502 503 504 505 506 507 |
if (likely(pid_alive(task))) { if (type != PIDTYPE_PID) task = task->group_leader; nr = pid_nr_ns(task->pids[type].pid, ns); } rcu_read_unlock(); return nr; |
2f2a3a46f
|
508 |
} |
52ee2dfdd
|
509 |
EXPORT_SYMBOL(__task_pid_nr_ns); |
2f2a3a46f
|
510 511 512 513 514 515 |
pid_t task_tgid_nr_ns(struct task_struct *tsk, struct pid_namespace *ns) { return pid_nr_ns(task_tgid(tsk), ns); } EXPORT_SYMBOL(task_tgid_nr_ns); |
61bce0f13
|
516 517 518 519 520 |
struct pid_namespace *task_active_pid_ns(struct task_struct *tsk) { return ns_of_pid(task_pid(tsk)); } EXPORT_SYMBOL_GPL(task_active_pid_ns); |
1da177e4c
|
521 |
/* |
025dfdafe
|
522 |
* Used by proc to find the first pid that is greater than or equal to nr. |
0804ef4b0
|
523 |
* |
e49859e71
|
524 |
* If there is a pid at nr this function is exactly the same as find_pid_ns. |
0804ef4b0
|
525 |
*/ |
198fe21b0
|
526 |
struct pid *find_ge_pid(int nr, struct pid_namespace *ns) |
0804ef4b0
|
527 528 529 530 |
{ struct pid *pid; do { |
198fe21b0
|
531 |
pid = find_pid_ns(nr, ns); |
0804ef4b0
|
532 533 |
if (pid) break; |
198fe21b0
|
534 |
nr = next_pidmap(ns, nr); |
0804ef4b0
|
535 536 537 538 539 540 |
} while (nr > 0); return pid; } /* |
1da177e4c
|
541 542 543 544 545 546 |
* The pid hash table is scaled according to the amount of memory in the * machine. From a minimum of 16 slots up to 4096 slots at one gigabyte or * more. */ void __init pidhash_init(void) { |
074b85175
|
547 |
unsigned int i, pidhash_size; |
1da177e4c
|
548 |
|
2c85f51d2
|
549 550 |
pid_hash = alloc_large_system_hash("PID", sizeof(*pid_hash), 0, 18, HASH_EARLY | HASH_SMALL, |
31fe62b95
|
551 552 |
&pidhash_shift, NULL, 0, 4096); |
074b85175
|
553 |
pidhash_size = 1U << pidhash_shift; |
1da177e4c
|
554 |
|
92476d7fc
|
555 556 |
for (i = 0; i < pidhash_size; i++) INIT_HLIST_HEAD(&pid_hash[i]); |
1da177e4c
|
557 558 559 560 |
} void __init pidmap_init(void) { |
c876ad768
|
561 562 |
/* Veryify no one has done anything silly */ BUILD_BUG_ON(PID_MAX_LIMIT >= PIDNS_HASH_ADDING); |
72680a191
|
563 564 565 566 567 568 569 |
/* bump default and minimum pid_max based on number of cpus */ pid_max = min(pid_max_max, max_t(int, pid_max, PIDS_PER_CPU_DEFAULT * num_possible_cpus())); pid_max_min = max_t(int, pid_max_min, PIDS_PER_CPU_MIN * num_possible_cpus()); pr_info("pid_max: default: %u minimum: %u ", pid_max, pid_max_min); |
61a58c6c2
|
570 |
init_pid_ns.pidmap[0].page = kzalloc(PAGE_SIZE, GFP_KERNEL); |
73b9ebfe1
|
571 |
/* Reserve PID 0. We never call free_pidmap(0) */ |
61a58c6c2
|
572 573 |
set_bit(0, init_pid_ns.pidmap[0].page); atomic_dec(&init_pid_ns.pidmap[0].nr_free); |
92476d7fc
|
574 |
|
74bd59bb3
|
575 576 |
init_pid_ns.pid_cachep = KMEM_CACHE(pid, SLAB_HWCACHE_ALIGN | SLAB_PANIC); |
1da177e4c
|
577 |
} |