fork.c 40.4 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
/*
 *  linux/kernel/fork.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 */

/*
 *  'fork.c' contains the help-routines for the 'fork' system call
 * (see also entry.S and others).
 * Fork is rather simple, once you get the hang of it, but the memory
 * management can be a bitch. See 'mm/memory.c': 'copy_page_range()'
 */

#include <linux/slab.h>
#include <linux/init.h>
#include <linux/unistd.h>
#include <linux/module.h>
#include <linux/vmalloc.h>
#include <linux/completion.h>
20
#include <linux/mnt_namespace.h>
Linus Torvalds's avatar
Linus Torvalds committed
21 22 23 24
#include <linux/personality.h>
#include <linux/mempolicy.h>
#include <linux/sem.h>
#include <linux/file.h>
Al Viro's avatar
Al Viro committed
25
#include <linux/fdtable.h>
26
#include <linux/iocontext.h>
Linus Torvalds's avatar
Linus Torvalds committed
27 28 29 30
#include <linux/key.h>
#include <linux/binfmts.h>
#include <linux/mman.h>
#include <linux/fs.h>
31
#include <linux/nsproxy.h>
32
#include <linux/capability.h>
Linus Torvalds's avatar
Linus Torvalds committed
33
#include <linux/cpu.h>
34
#include <linux/cgroup.h>
Linus Torvalds's avatar
Linus Torvalds committed
35 36 37 38 39
#include <linux/security.h>
#include <linux/swap.h>
#include <linux/syscalls.h>
#include <linux/jiffies.h>
#include <linux/futex.h>
40
#include <linux/task_io_accounting_ops.h>
41
#include <linux/rcupdate.h>
Linus Torvalds's avatar
Linus Torvalds committed
42 43 44
#include <linux/ptrace.h>
#include <linux/mount.h>
#include <linux/audit.h>
45
#include <linux/memcontrol.h>
Linus Torvalds's avatar
Linus Torvalds committed
46 47 48
#include <linux/profile.h>
#include <linux/rmap.h>
#include <linux/acct.h>
49
#include <linux/tsacct_kern.h>
50
#include <linux/cn_proc.h>
51
#include <linux/freezer.h>
52
#include <linux/delayacct.h>
53
#include <linux/taskstats_kern.h>
54
#include <linux/random.h>
55
#include <linux/tty.h>
56
#include <linux/proc_fs.h>
57
#include <linux/blkdev.h>
Linus Torvalds's avatar
Linus Torvalds committed
58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75

#include <asm/pgtable.h>
#include <asm/pgalloc.h>
#include <asm/uaccess.h>
#include <asm/mmu_context.h>
#include <asm/cacheflush.h>
#include <asm/tlbflush.h>

/*
 * Protected counters by write_lock_irq(&tasklist_lock)
 */
unsigned long total_forks;	/* Handle normal Linux uptimes. */
int nr_threads; 		/* The idle threads do not count.. */

int max_threads;		/* tunable limit on nr_threads */

DEFINE_PER_CPU(unsigned long, process_counts) = 0;

76
__cacheline_aligned DEFINE_RWLOCK(tasklist_lock);  /* outer */
Linus Torvalds's avatar
Linus Torvalds committed
77 78 79 80 81 82 83 84 85 86 87 88 89 90 91

int nr_processes(void)
{
	int cpu;
	int total = 0;

	for_each_online_cpu(cpu)
		total += per_cpu(process_counts, cpu);

	return total;
}

#ifndef __HAVE_ARCH_TASK_STRUCT_ALLOCATOR
# define alloc_task_struct()	kmem_cache_alloc(task_struct_cachep, GFP_KERNEL)
# define free_task_struct(tsk)	kmem_cache_free(task_struct_cachep, (tsk))
92
static struct kmem_cache *task_struct_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
93 94 95
#endif

/* SLAB cache for signal_struct structures (tsk->signal) */
96
static struct kmem_cache *signal_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
97 98

/* SLAB cache for sighand_struct structures (tsk->sighand) */
99
struct kmem_cache *sighand_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
100 101

/* SLAB cache for files_struct structures (tsk->files) */
102
struct kmem_cache *files_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
103 104

/* SLAB cache for fs_struct structures (tsk->fs) */
105
struct kmem_cache *fs_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
106 107

/* SLAB cache for vm_area_struct structures */
108
struct kmem_cache *vm_area_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
109 110

/* SLAB cache for mm_struct structures (tsk->mm) */
111
static struct kmem_cache *mm_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
112 113 114

void free_task(struct task_struct *tsk)
{
115
	prop_local_destroy_single(&tsk->dirties);
Roman Zippel's avatar
Roman Zippel committed
116
	free_thread_info(tsk->stack);
117
	rt_mutex_debug_task_free(tsk);
Linus Torvalds's avatar
Linus Torvalds committed
118 119 120 121
	free_task_struct(tsk);
}
EXPORT_SYMBOL(free_task);

122
void __put_task_struct(struct task_struct *tsk)
Linus Torvalds's avatar
Linus Torvalds committed
123
{
Eugene Teo's avatar
Eugene Teo committed
124
	WARN_ON(!tsk->exit_state);
Linus Torvalds's avatar
Linus Torvalds committed
125 126 127 128 129 130
	WARN_ON(atomic_read(&tsk->usage));
	WARN_ON(tsk == current);

	security_task_free(tsk);
	free_uid(tsk->user);
	put_group_info(tsk->group_info);
131
	delayacct_tsk_free(tsk);
Linus Torvalds's avatar
Linus Torvalds committed
132 133 134 135 136

	if (!profile_handoff_task(tsk))
		free_task(tsk);
}

Suresh Siddha's avatar
Suresh Siddha committed
137 138 139 140 141 142 143
/*
 * macro override instead of weak attribute alias, to workaround
 * gcc 4.1.0 and 4.1.1 bugs with weak attribute and empty functions.
 */
#ifndef arch_task_cache_init
#define arch_task_cache_init()
#endif
144

Linus Torvalds's avatar
Linus Torvalds committed
145 146 147 148 149 150 151 152 153
void __init fork_init(unsigned long mempages)
{
#ifndef __HAVE_ARCH_TASK_STRUCT_ALLOCATOR
#ifndef ARCH_MIN_TASKALIGN
#define ARCH_MIN_TASKALIGN	L1_CACHE_BYTES
#endif
	/* create a slab on which task_structs can be allocated */
	task_struct_cachep =
		kmem_cache_create("task_struct", sizeof(struct task_struct),
154
			ARCH_MIN_TASKALIGN, SLAB_PANIC, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
155 156
#endif

157 158 159
	/* do the arch specific task caches init */
	arch_task_cache_init();

Linus Torvalds's avatar
Linus Torvalds committed
160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178
	/*
	 * The default maximum number of threads is set to a safe
	 * value: the thread structures can take up at most half
	 * of memory.
	 */
	max_threads = mempages / (8 * THREAD_SIZE / PAGE_SIZE);

	/*
	 * we need to allow at least 20 threads to boot a system
	 */
	if(max_threads < 20)
		max_threads = 20;

	init_task.signal->rlim[RLIMIT_NPROC].rlim_cur = max_threads/2;
	init_task.signal->rlim[RLIMIT_NPROC].rlim_max = max_threads/2;
	init_task.signal->rlim[RLIMIT_SIGPENDING] =
		init_task.signal->rlim[RLIMIT_NPROC];
}

179 180 181 182 183 184 185
int __attribute__((weak)) arch_dup_task_struct(struct task_struct *dst,
					       struct task_struct *src)
{
	*dst = *src;
	return 0;
}

Linus Torvalds's avatar
Linus Torvalds committed
186 187 188 189
static struct task_struct *dup_task_struct(struct task_struct *orig)
{
	struct task_struct *tsk;
	struct thread_info *ti;
190
	int err;
Linus Torvalds's avatar
Linus Torvalds committed
191 192 193 194 195 196 197 198 199 200 201 202 203

	prepare_to_copy(orig);

	tsk = alloc_task_struct();
	if (!tsk)
		return NULL;

	ti = alloc_thread_info(tsk);
	if (!ti) {
		free_task_struct(tsk);
		return NULL;
	}

204 205 206 207
 	err = arch_dup_task_struct(tsk, orig);
	if (err)
		goto out;

Roman Zippel's avatar
Roman Zippel committed
208
	tsk->stack = ti;
209 210

	err = prop_local_init_single(&tsk->dirties);
211 212
	if (err)
		goto out;
213

214
	setup_thread_stack(tsk, orig);
Linus Torvalds's avatar
Linus Torvalds committed
215

216 217 218 219
#ifdef CONFIG_CC_STACKPROTECTOR
	tsk->stack_canary = get_random_int();
#endif

Linus Torvalds's avatar
Linus Torvalds committed
220 221
	/* One for us, one for whoever does the "release_task()" (usually parent) */
	atomic_set(&tsk->usage,2);
222
	atomic_set(&tsk->fs_excl, 0);
223
#ifdef CONFIG_BLK_DEV_IO_TRACE
224
	tsk->btrace_seq = 0;
225
#endif
226
	tsk->splice_pipe = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
227
	return tsk;
228 229 230 231 232

out:
	free_thread_info(ti);
	free_task_struct(tsk);
	return NULL;
Linus Torvalds's avatar
Linus Torvalds committed
233 234 235
}

#ifdef CONFIG_MMU
Alexey Dobriyan's avatar
Alexey Dobriyan committed
236
static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm)
Linus Torvalds's avatar
Linus Torvalds committed
237
{
238
	struct vm_area_struct *mpnt, *tmp, **pprev;
Linus Torvalds's avatar
Linus Torvalds committed
239 240 241 242 243 244
	struct rb_node **rb_link, *rb_parent;
	int retval;
	unsigned long charge;
	struct mempolicy *pol;

	down_write(&oldmm->mmap_sem);
245
	flush_cache_dup_mm(oldmm);
246 247 248 249
	/*
	 * Not linked in yet - no deadlock potential:
	 */
	down_write_nested(&mm->mmap_sem, SINGLE_DEPTH_NESTING);
250

Linus Torvalds's avatar
Linus Torvalds committed
251 252 253 254
	mm->locked_vm = 0;
	mm->mmap = NULL;
	mm->mmap_cache = NULL;
	mm->free_area_cache = oldmm->mmap_base;
255
	mm->cached_hole_size = ~0UL;
Linus Torvalds's avatar
Linus Torvalds committed
256 257 258 259 260 261 262
	mm->map_count = 0;
	cpus_clear(mm->cpu_vm_mask);
	mm->mm_rb = RB_ROOT;
	rb_link = &mm->mm_rb.rb_node;
	rb_parent = NULL;
	pprev = &mm->mmap;

263
	for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) {
Linus Torvalds's avatar
Linus Torvalds committed
264 265 266
		struct file *file;

		if (mpnt->vm_flags & VM_DONTCOPY) {
267 268
			long pages = vma_pages(mpnt);
			mm->total_vm -= pages;
269
			vm_stat_account(mm, mpnt->vm_flags, mpnt->vm_file,
270
								-pages);
Linus Torvalds's avatar
Linus Torvalds committed
271 272 273 274 275 276 277 278 279
			continue;
		}
		charge = 0;
		if (mpnt->vm_flags & VM_ACCOUNT) {
			unsigned int len = (mpnt->vm_end - mpnt->vm_start) >> PAGE_SHIFT;
			if (security_vm_enough_memory(len))
				goto fail_nomem;
			charge = len;
		}
280
		tmp = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL);
Linus Torvalds's avatar
Linus Torvalds committed
281 282 283
		if (!tmp)
			goto fail_nomem;
		*tmp = *mpnt;
284
		pol = mpol_dup(vma_policy(mpnt));
Linus Torvalds's avatar
Linus Torvalds committed
285 286 287 288 289 290 291 292 293 294
		retval = PTR_ERR(pol);
		if (IS_ERR(pol))
			goto fail_nomem_policy;
		vma_set_policy(tmp, pol);
		tmp->vm_flags &= ~VM_LOCKED;
		tmp->vm_mm = mm;
		tmp->vm_next = NULL;
		anon_vma_link(tmp);
		file = tmp->vm_file;
		if (file) {
295
			struct inode *inode = file->f_path.dentry->d_inode;
Linus Torvalds's avatar
Linus Torvalds committed
296 297 298
			get_file(file);
			if (tmp->vm_flags & VM_DENYWRITE)
				atomic_dec(&inode->i_writecount);
Daniel Walker's avatar
Daniel Walker committed
299

Linus Torvalds's avatar
Linus Torvalds committed
300 301 302 303 304 305 306 307 308 309
			/* insert tmp into the share list, just after mpnt */
			spin_lock(&file->f_mapping->i_mmap_lock);
			tmp->vm_truncate_count = mpnt->vm_truncate_count;
			flush_dcache_mmap_lock(file->f_mapping);
			vma_prio_tree_add(tmp, mpnt);
			flush_dcache_mmap_unlock(file->f_mapping);
			spin_unlock(&file->f_mapping->i_mmap_lock);
		}

		/*
310
		 * Link in the new vma and copy the page table entries.
Linus Torvalds's avatar
Linus Torvalds committed
311 312 313 314 315 316 317 318 319
		 */
		*pprev = tmp;
		pprev = &tmp->vm_next;

		__vma_link_rb(mm, tmp, rb_link, rb_parent);
		rb_link = &tmp->vm_rb.rb_right;
		rb_parent = &tmp->vm_rb;

		mm->map_count++;
320
		retval = copy_page_range(mm, oldmm, mpnt);
Linus Torvalds's avatar
Linus Torvalds committed
321 322 323 324 325 326 327

		if (tmp->vm_ops && tmp->vm_ops->open)
			tmp->vm_ops->open(tmp);

		if (retval)
			goto out;
	}
328 329
	/* a new mm has just been created */
	arch_dup_mmap(oldmm, mm);
Linus Torvalds's avatar
Linus Torvalds committed
330 331
	retval = 0;
out:
332
	up_write(&mm->mmap_sem);
333
	flush_tlb_mm(oldmm);
Linus Torvalds's avatar
Linus Torvalds committed
334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353
	up_write(&oldmm->mmap_sem);
	return retval;
fail_nomem_policy:
	kmem_cache_free(vm_area_cachep, tmp);
fail_nomem:
	retval = -ENOMEM;
	vm_unacct_memory(charge);
	goto out;
}

static inline int mm_alloc_pgd(struct mm_struct * mm)
{
	mm->pgd = pgd_alloc(mm);
	if (unlikely(!mm->pgd))
		return -ENOMEM;
	return 0;
}

static inline void mm_free_pgd(struct mm_struct * mm)
{
354
	pgd_free(mm, mm->pgd);
Linus Torvalds's avatar
Linus Torvalds committed
355 356 357 358 359 360 361
}
#else
#define dup_mmap(mm, oldmm)	(0)
#define mm_alloc_pgd(mm)	(0)
#define mm_free_pgd(mm)
#endif /* CONFIG_MMU */

Daniel Walker's avatar
Daniel Walker committed
362
__cacheline_aligned_in_smp DEFINE_SPINLOCK(mmlist_lock);
Linus Torvalds's avatar
Linus Torvalds committed
363

364
#define allocate_mm()	(kmem_cache_alloc(mm_cachep, GFP_KERNEL))
Linus Torvalds's avatar
Linus Torvalds committed
365 366 367 368
#define free_mm(mm)	(kmem_cache_free(mm_cachep, (mm)))

#include <linux/init_task.h>

369
static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p)
Linus Torvalds's avatar
Linus Torvalds committed
370 371 372 373 374
{
	atomic_set(&mm->mm_users, 1);
	atomic_set(&mm->mm_count, 1);
	init_rwsem(&mm->mmap_sem);
	INIT_LIST_HEAD(&mm->mmlist);
375 376
	mm->flags = (current->mm) ? current->mm->flags
				  : MMF_DUMP_FILTER_DEFAULT;
Linus Torvalds's avatar
Linus Torvalds committed
377 378
	mm->core_waiters = 0;
	mm->nr_ptes = 0;
379
	set_mm_counter(mm, file_rss, 0);
380
	set_mm_counter(mm, anon_rss, 0);
Linus Torvalds's avatar
Linus Torvalds committed
381 382 383 384
	spin_lock_init(&mm->page_table_lock);
	rwlock_init(&mm->ioctx_list_lock);
	mm->ioctx_list = NULL;
	mm->free_area_cache = TASK_UNMAPPED_BASE;
385
	mm->cached_hole_size = ~0UL;
386
	mm_init_owner(mm, p);
Linus Torvalds's avatar
Linus Torvalds committed
387 388 389 390 391

	if (likely(!mm_alloc_pgd(mm))) {
		mm->def_flags = 0;
		return mm;
	}
392

Linus Torvalds's avatar
Linus Torvalds committed
393 394 395 396 397 398 399 400 401 402 403 404 405 406
	free_mm(mm);
	return NULL;
}

/*
 * Allocate and initialize an mm_struct.
 */
struct mm_struct * mm_alloc(void)
{
	struct mm_struct * mm;

	mm = allocate_mm();
	if (mm) {
		memset(mm, 0, sizeof(*mm));
407
		mm = mm_init(mm, current);
Linus Torvalds's avatar
Linus Torvalds committed
408 409 410 411 412 413 414 415 416
	}
	return mm;
}

/*
 * Called when the last reference to the mm
 * is dropped: either by a lazy thread or by
 * mmput. Free the page directory and the mm.
 */
417
void __mmdrop(struct mm_struct *mm)
Linus Torvalds's avatar
Linus Torvalds committed
418 419 420 421 422 423
{
	BUG_ON(mm == &init_mm);
	mm_free_pgd(mm);
	destroy_context(mm);
	free_mm(mm);
}
424
EXPORT_SYMBOL_GPL(__mmdrop);
Linus Torvalds's avatar
Linus Torvalds committed
425 426 427 428 429 430

/*
 * Decrement the use count and release all resources for an mm.
 */
void mmput(struct mm_struct *mm)
{
Andrew Morton's avatar
Andrew Morton committed
431 432
	might_sleep();

Linus Torvalds's avatar
Linus Torvalds committed
433 434 435
	if (atomic_dec_and_test(&mm->mm_users)) {
		exit_aio(mm);
		exit_mmap(mm);
Matt Helsley's avatar
Matt Helsley committed
436
		set_mm_exe_file(mm, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498
		if (!list_empty(&mm->mmlist)) {
			spin_lock(&mmlist_lock);
			list_del(&mm->mmlist);
			spin_unlock(&mmlist_lock);
		}
		put_swap_token(mm);
		mmdrop(mm);
	}
}
EXPORT_SYMBOL_GPL(mmput);

/**
 * get_task_mm - acquire a reference to the task's mm
 *
 * Returns %NULL if the task has no mm.  Checks PF_BORROWED_MM (meaning
 * this kernel workthread has transiently adopted a user mm with use_mm,
 * to do its AIO) is not set and if so returns a reference to it, after
 * bumping up the use count.  User must release the mm via mmput()
 * after use.  Typically used by /proc and ptrace.
 */
struct mm_struct *get_task_mm(struct task_struct *task)
{
	struct mm_struct *mm;

	task_lock(task);
	mm = task->mm;
	if (mm) {
		if (task->flags & PF_BORROWED_MM)
			mm = NULL;
		else
			atomic_inc(&mm->mm_users);
	}
	task_unlock(task);
	return mm;
}
EXPORT_SYMBOL_GPL(get_task_mm);

/* Please note the differences between mmput and mm_release.
 * mmput is called whenever we stop holding onto a mm_struct,
 * error success whatever.
 *
 * mm_release is called after a mm_struct has been removed
 * from the current process.
 *
 * This difference is important for error handling, when we
 * only half set up a mm_struct for a new process and need to restore
 * the old one.  Because we mmput the new mm_struct before
 * restoring the old one. . .
 * Eric Biederman 10 January 1998
 */
void mm_release(struct task_struct *tsk, struct mm_struct *mm)
{
	struct completion *vfork_done = tsk->vfork_done;

	/* Get rid of any cached register state */
	deactivate_mm(tsk, mm);

	/* notify parent sleeping on vfork() */
	if (vfork_done) {
		tsk->vfork_done = NULL;
		complete(vfork_done);
	}
499 500 501 502 503 504 505 506 507 508

	/*
	 * If we're exiting normally, clear a user-space tid field if
	 * requested.  We leave this alone when dying by signal, to leave
	 * the value intact in a core dump, and to save the unnecessary
	 * trouble otherwise.  Userland only wants this done for a sys_exit.
	 */
	if (tsk->clear_child_tid
	    && !(tsk->flags & PF_SIGNALED)
	    && atomic_read(&mm->mm_users) > 1) {
Linus Torvalds's avatar
Linus Torvalds committed
509 510 511 512 513 514 515 516 517 518 519 520
		u32 __user * tidptr = tsk->clear_child_tid;
		tsk->clear_child_tid = NULL;

		/*
		 * We don't check the error code - if userspace has
		 * not set up a proper pointer then tough luck.
		 */
		put_user(0, tidptr);
		sys_futex(tidptr, FUTEX_WAKE, 1, NULL, NULL, 0);
	}
}

521 522 523 524
/*
 * Allocate a new mm structure and copy contents from the
 * mm structure of the passed in task structure.
 */
525
struct mm_struct *dup_mm(struct task_struct *tsk)
526 527 528 529 530 531 532 533 534 535 536 537 538
{
	struct mm_struct *mm, *oldmm = current->mm;
	int err;

	if (!oldmm)
		return NULL;

	mm = allocate_mm();
	if (!mm)
		goto fail_nomem;

	memcpy(mm, oldmm, sizeof(*mm));

539 540 541 542
	/* Initializing for Swap token stuff */
	mm->token_priority = 0;
	mm->last_interval = 0;

543
	if (!mm_init(mm, tsk))
544 545 546 547 548
		goto fail_nomem;

	if (init_new_context(tsk, mm))
		goto fail_nocontext;

Matt Helsley's avatar
Matt Helsley committed
549 550
	dup_mm_exe_file(oldmm, mm);

551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575
	err = dup_mmap(mm, oldmm);
	if (err)
		goto free_pt;

	mm->hiwater_rss = get_mm_rss(mm);
	mm->hiwater_vm = mm->total_vm;

	return mm;

free_pt:
	mmput(mm);

fail_nomem:
	return NULL;

fail_nocontext:
	/*
	 * If init_new_context() failed, we cannot use mmput() to free the mm
	 * because it calls destroy_context()
	 */
	mm_free_pgd(mm);
	free_mm(mm);
	return NULL;
}

Linus Torvalds's avatar
Linus Torvalds committed
576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602
static int copy_mm(unsigned long clone_flags, struct task_struct * tsk)
{
	struct mm_struct * mm, *oldmm;
	int retval;

	tsk->min_flt = tsk->maj_flt = 0;
	tsk->nvcsw = tsk->nivcsw = 0;

	tsk->mm = NULL;
	tsk->active_mm = NULL;

	/*
	 * Are we cloning a kernel thread?
	 *
	 * We need to steal a active VM for that..
	 */
	oldmm = current->mm;
	if (!oldmm)
		return 0;

	if (clone_flags & CLONE_VM) {
		atomic_inc(&oldmm->mm_users);
		mm = oldmm;
		goto good_mm;
	}

	retval = -ENOMEM;
603
	mm = dup_mm(tsk);
Linus Torvalds's avatar
Linus Torvalds committed
604 605 606 607
	if (!mm)
		goto fail_nomem;

good_mm:
608 609 610 611
	/* Initializing for Swap token stuff */
	mm->token_priority = 0;
	mm->last_interval = 0;

Linus Torvalds's avatar
Linus Torvalds committed
612 613 614 615 616 617 618 619
	tsk->mm = mm;
	tsk->active_mm = mm;
	return 0;

fail_nomem:
	return retval;
}

Alexey Dobriyan's avatar
Alexey Dobriyan committed
620
static struct fs_struct *__copy_fs_struct(struct fs_struct *old)
Linus Torvalds's avatar
Linus Torvalds committed
621 622 623 624 625 626 627 628
{
	struct fs_struct *fs = kmem_cache_alloc(fs_cachep, GFP_KERNEL);
	/* We don't need to lock fs - think why ;-) */
	if (fs) {
		atomic_set(&fs->count, 1);
		rwlock_init(&fs->lock);
		fs->umask = old->umask;
		read_lock(&old->lock);
Jan Blunck's avatar
Jan Blunck committed
629 630 631 632 633 634 635
		fs->root = old->root;
		path_get(&old->root);
		fs->pwd = old->pwd;
		path_get(&old->pwd);
		if (old->altroot.dentry) {
			fs->altroot = old->altroot;
			path_get(&old->altroot);
Linus Torvalds's avatar
Linus Torvalds committed
636
		} else {
Jan Blunck's avatar
Jan Blunck committed
637 638
			fs->altroot.mnt = NULL;
			fs->altroot.dentry = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
639 640 641 642 643 644 645 646 647 648 649 650 651
		}
		read_unlock(&old->lock);
	}
	return fs;
}

struct fs_struct *copy_fs_struct(struct fs_struct *old)
{
	return __copy_fs_struct(old);
}

EXPORT_SYMBOL_GPL(copy_fs_struct);

Alexey Dobriyan's avatar
Alexey Dobriyan committed
652
static int copy_fs(unsigned long clone_flags, struct task_struct *tsk)
Linus Torvalds's avatar
Linus Torvalds committed
653 654 655 656 657 658 659 660 661 662 663
{
	if (clone_flags & CLONE_FS) {
		atomic_inc(&current->fs->count);
		return 0;
	}
	tsk->fs = __copy_fs_struct(current->fs);
	if (!tsk->fs)
		return -ENOMEM;
	return 0;
}

664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690
static int copy_files(unsigned long clone_flags, struct task_struct * tsk)
{
	struct files_struct *oldf, *newf;
	int error = 0;

	/*
	 * A background process may not have any files ...
	 */
	oldf = current->files;
	if (!oldf)
		goto out;

	if (clone_flags & CLONE_FILES) {
		atomic_inc(&oldf->count);
		goto out;
	}

	newf = dup_fd(oldf, &error);
	if (!newf)
		goto out;

	tsk->files = newf;
	error = 0;
out:
	return error;
}

691
static int copy_io(unsigned long clone_flags, struct task_struct *tsk)
692 693 694 695 696 697
{
#ifdef CONFIG_BLOCK
	struct io_context *ioc = current->io_context;

	if (!ioc)
		return 0;
698 699 700 701 702 703 704 705
	/*
	 * Share io context with parent, if CLONE_IO is set
	 */
	if (clone_flags & CLONE_IO) {
		tsk->io_context = ioc_task_link(ioc);
		if (unlikely(!tsk->io_context))
			return -ENOMEM;
	} else if (ioprio_valid(ioc->ioprio)) {
706 707 708 709 710 711 712 713 714 715
		tsk->io_context = alloc_io_context(GFP_KERNEL, -1);
		if (unlikely(!tsk->io_context))
			return -ENOMEM;

		tsk->io_context->ioprio = ioc->ioprio;
	}
#endif
	return 0;
}

Alexey Dobriyan's avatar
Alexey Dobriyan committed
716
static int copy_sighand(unsigned long clone_flags, struct task_struct *tsk)
Linus Torvalds's avatar
Linus Torvalds committed
717 718 719 720 721 722 723 724
{
	struct sighand_struct *sig;

	if (clone_flags & (CLONE_SIGHAND | CLONE_THREAD)) {
		atomic_inc(&current->sighand->count);
		return 0;
	}
	sig = kmem_cache_alloc(sighand_cachep, GFP_KERNEL);
Ingo Molnar's avatar
Ingo Molnar committed
725
	rcu_assign_pointer(tsk->sighand, sig);
Linus Torvalds's avatar
Linus Torvalds committed
726 727 728 729 730 731 732
	if (!sig)
		return -ENOMEM;
	atomic_set(&sig->count, 1);
	memcpy(sig->action, current->sighand->action, sizeof(sig->action));
	return 0;
}

733
void __cleanup_sighand(struct sighand_struct *sighand)
734 735 736 737 738
{
	if (atomic_dec_and_test(&sighand->count))
		kmem_cache_free(sighand_cachep, sighand);
}

Alexey Dobriyan's avatar
Alexey Dobriyan committed
739
static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
Linus Torvalds's avatar
Linus Torvalds committed
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
{
	struct signal_struct *sig;
	int ret;

	if (clone_flags & CLONE_THREAD) {
		atomic_inc(&current->signal->count);
		atomic_inc(&current->signal->live);
		return 0;
	}
	sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL);
	tsk->signal = sig;
	if (!sig)
		return -ENOMEM;

	ret = copy_thread_group_keys(tsk);
	if (ret < 0) {
		kmem_cache_free(signal_cachep, sig);
		return ret;
	}

	atomic_set(&sig->count, 1);
	atomic_set(&sig->live, 1);
	init_waitqueue_head(&sig->wait_chldexit);
	sig->flags = 0;
	sig->group_exit_code = 0;
	sig->group_exit_task = NULL;
	sig->group_stop_count = 0;
767
	sig->curr_target = tsk;
Linus Torvalds's avatar
Linus Torvalds committed
768 769 770
	init_sigpending(&sig->shared_pending);
	INIT_LIST_HEAD(&sig->posix_timers);

771
	hrtimer_init(&sig->real_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
772
	sig->it_real_incr.tv64 = 0;
Linus Torvalds's avatar
Linus Torvalds committed
773 774 775 776 777 778 779 780
	sig->real_timer.function = it_real_fn;

	sig->it_virt_expires = cputime_zero;
	sig->it_virt_incr = cputime_zero;
	sig->it_prof_expires = cputime_zero;
	sig->it_prof_incr = cputime_zero;

	sig->leader = 0;	/* session leadership doesn't inherit */
781
	sig->tty_old_pgrp = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
782 783

	sig->utime = sig->stime = sig->cutime = sig->cstime = cputime_zero;
784 785
	sig->gtime = cputime_zero;
	sig->cgtime = cputime_zero;
Linus Torvalds's avatar
Linus Torvalds committed
786 787
	sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0;
	sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0;
788
	sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0;
789
	sig->sum_sched_runtime = 0;
Linus Torvalds's avatar
Linus Torvalds committed
790 791 792
	INIT_LIST_HEAD(&sig->cpu_timers[0]);
	INIT_LIST_HEAD(&sig->cpu_timers[1]);
	INIT_LIST_HEAD(&sig->cpu_timers[2]);
793
	taskstats_tgid_init(sig);
Linus Torvalds's avatar
Linus Torvalds committed
794 795 796 797 798 799 800 801 802 803 804 805 806

	task_lock(current->group_leader);
	memcpy(sig->rlim, current->signal->rlim, sizeof sig->rlim);
	task_unlock(current->group_leader);

	if (sig->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY) {
		/*
		 * New sole thread in the process gets an expiry time
		 * of the whole CPU time limit.
		 */
		tsk->it_prof_expires =
			secs_to_cputime(sig->rlim[RLIMIT_CPU].rlim_cur);
	}
807
	acct_init_pacct(&sig->pacct);
Linus Torvalds's avatar
Linus Torvalds committed
808

809 810
	tty_audit_fork(sig);

Linus Torvalds's avatar
Linus Torvalds committed
811 812 813
	return 0;
}

814 815 816 817 818 819
void __cleanup_signal(struct signal_struct *sig)
{
	exit_thread_group_keys(sig);
	kmem_cache_free(signal_cachep, sig);
}

Alexey Dobriyan's avatar
Alexey Dobriyan committed
820
static void cleanup_signal(struct task_struct *tsk)
821 822 823 824 825 826 827 828 829
{
	struct signal_struct *sig = tsk->signal;

	atomic_dec(&sig->live);

	if (atomic_dec_and_test(&sig->count))
		__cleanup_signal(sig);
}

Alexey Dobriyan's avatar
Alexey Dobriyan committed
830
static void copy_flags(unsigned long clone_flags, struct task_struct *p)
Linus Torvalds's avatar
Linus Torvalds committed
831 832 833
{
	unsigned long new_flags = p->flags;

834
	new_flags &= ~PF_SUPERPRIV;
Linus Torvalds's avatar
Linus Torvalds committed
835 836 837 838
	new_flags |= PF_FORKNOEXEC;
	if (!(clone_flags & CLONE_PTRACE))
		p->ptrace = 0;
	p->flags = new_flags;
839
	clear_freeze_flag(p);
Linus Torvalds's avatar
Linus Torvalds committed
840 841 842 843 844 845
}

asmlinkage long sys_set_tid_address(int __user *tidptr)
{
	current->clear_child_tid = tidptr;

846
	return task_pid_vnr(current);
Linus Torvalds's avatar
Linus Torvalds committed
847 848
}

Alexey Dobriyan's avatar
Alexey Dobriyan committed
849
static void rt_mutex_init_task(struct task_struct *p)
850 851
{
	spin_lock_init(&p->pi_lock);
852
#ifdef CONFIG_RT_MUTEXES
853 854 855 856 857
	plist_head_init(&p->pi_waiters, &p->pi_lock);
	p->pi_blocked_on = NULL;
#endif
}

858 859 860 861 862 863 864
#ifdef CONFIG_MM_OWNER
void mm_init_owner(struct mm_struct *mm, struct task_struct *p)
{
	mm->owner = p;
}
#endif /* CONFIG_MM_OWNER */

Linus Torvalds's avatar
Linus Torvalds committed
865 866 867 868 869 870 871 872
/*
 * This creates a new process as a copy of the old one,
 * but does not actually start it yet.
 *
 * It copies the registers, and all the appropriate
 * parts of the process environment (as per the clone
 * flags). The actual kick-off is left to the caller.
 */
873 874 875 876 877
static struct task_struct *copy_process(unsigned long clone_flags,
					unsigned long stack_start,
					struct pt_regs *regs,
					unsigned long stack_size,
					int __user *child_tidptr,
878
					struct pid *pid)
Linus Torvalds's avatar
Linus Torvalds committed
879 880
{
	int retval;
881
	struct task_struct *p;
882
	int cgroup_callbacks_done = 0;
Linus Torvalds's avatar
Linus Torvalds committed
883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910

	if ((clone_flags & (CLONE_NEWNS|CLONE_FS)) == (CLONE_NEWNS|CLONE_FS))
		return ERR_PTR(-EINVAL);

	/*
	 * Thread groups must share signals as well, and detached threads
	 * can only be started up within the thread group.
	 */
	if ((clone_flags & CLONE_THREAD) && !(clone_flags & CLONE_SIGHAND))
		return ERR_PTR(-EINVAL);

	/*
	 * Shared signal handlers imply shared VM. By way of the above,
	 * thread groups also imply shared VM. Blocking this case allows
	 * for various simplifications in other code.
	 */
	if ((clone_flags & CLONE_SIGHAND) && !(clone_flags & CLONE_VM))
		return ERR_PTR(-EINVAL);

	retval = security_task_create(clone_flags);
	if (retval)
		goto fork_out;

	retval = -ENOMEM;
	p = dup_task_struct(current);
	if (!p)
		goto fork_out;

911 912
	rt_mutex_init_task(p);

913
#ifdef CONFIG_PROVE_LOCKING
914 915 916
	DEBUG_LOCKS_WARN_ON(!p->hardirqs_enabled);
	DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled);
#endif
Linus Torvalds's avatar
Linus Torvalds committed
917 918 919 920
	retval = -EAGAIN;
	if (atomic_read(&p->user->processes) >=
			p->signal->rlim[RLIMIT_NPROC].rlim_cur) {
		if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RESOURCE) &&
921
		    p->user != current->nsproxy->user_ns->root_user)
Linus Torvalds's avatar
Linus Torvalds committed
922 923 924 925 926 927 928 929 930 931 932 933 934 935 936
			goto bad_fork_free;
	}

	atomic_inc(&p->user->__count);
	atomic_inc(&p->user->processes);
	get_group_info(p->group_info);

	/*
	 * If multiple threads are within copy_process(), then this check
	 * triggers too late. This doesn't hurt, the check is only there
	 * to stop root fork bombs.
	 */
	if (nr_threads >= max_threads)
		goto bad_fork_cleanup_count;

937
	if (!try_module_get(task_thread_info(p)->exec_domain->module))
Linus Torvalds's avatar
Linus Torvalds committed
938 939 940 941 942 943
		goto bad_fork_cleanup_count;

	if (p->binfmt && !try_module_get(p->binfmt->module))
		goto bad_fork_cleanup_put_domain;

	p->did_exec = 0;
944
	delayacct_tsk_init(p);	/* Must remain after dup_task_struct() */
Linus Torvalds's avatar
Linus Torvalds committed
945 946 947
	copy_flags(clone_flags, p);
	INIT_LIST_HEAD(&p->children);
	INIT_LIST_HEAD(&p->sibling);
948 949 950 951
#ifdef CONFIG_PREEMPT_RCU
	p->rcu_read_lock_nesting = 0;
	p->rcu_flipctr_idx = 0;
#endif /* #ifdef CONFIG_PREEMPT_RCU */
Linus Torvalds's avatar
Linus Torvalds committed
952 953 954 955 956 957 958 959
	p->vfork_done = NULL;
	spin_lock_init(&p->alloc_lock);

	clear_tsk_thread_flag(p, TIF_SIGPENDING);
	init_sigpending(&p->pending);

	p->utime = cputime_zero;
	p->stime = cputime_zero;
960
	p->gtime = cputime_zero;
961 962
	p->utimescaled = cputime_zero;
	p->stimescaled = cputime_zero;
963
	p->prev_utime = cputime_zero;
964
	p->prev_stime = cputime_zero;
965

966 967 968 969 970
#ifdef CONFIG_DETECT_SOFTLOCKUP
	p->last_switch_count = 0;
	p->last_switch_timestamp = 0;
#endif

971
#ifdef CONFIG_TASK_XACCT
Linus Torvalds's avatar
Linus Torvalds committed
972 973 974 975
	p->rchar = 0;		/* I/O counter: bytes read */
	p->wchar = 0;		/* I/O counter: bytes written */
	p->syscr = 0;		/* I/O counter: read syscalls */
	p->syscw = 0;		/* I/O counter: write syscalls */
976
#endif
977
	task_io_accounting_init(p);
Linus Torvalds's avatar
Linus Torvalds committed
978 979
	acct_clear_integrals(p);

Daniel Walker's avatar
Daniel Walker committed
980
	p->it_virt_expires = cputime_zero;
Linus Torvalds's avatar
Linus Torvalds committed
981
	p->it_prof_expires = cputime_zero;
Daniel Walker's avatar
Daniel Walker committed
982 983 984 985
	p->it_sched_expires = 0;
	INIT_LIST_HEAD(&p->cpu_timers[0]);
	INIT_LIST_HEAD(&p->cpu_timers[1]);
	INIT_LIST_HEAD(&p->cpu_timers[2]);
Linus Torvalds's avatar
Linus Torvalds committed
986 987 988

	p->lock_depth = -1;		/* -1 = no lock */
	do_posix_clock_monotonic_gettime(&p->start_time);
989 990
	p->real_start_time = p->start_time;
	monotonic_to_bootbased(&p->real_start_time);
991
#ifdef CONFIG_SECURITY
Linus Torvalds's avatar
Linus Torvalds committed
992
	p->security = NULL;
993
#endif
994
	p->cap_bset = current->cap_bset;
Linus Torvalds's avatar
Linus Torvalds committed
995 996
	p->io_context = NULL;
	p->audit_context = NULL;
997
	cgroup_fork(p);
Linus Torvalds's avatar
Linus Torvalds committed
998
#ifdef CONFIG_NUMA
999
	p->mempolicy = mpol_dup(p->mempolicy);
Linus Torvalds's avatar
Linus Torvalds committed
1000 1001 1002
 	if (IS_ERR(p->mempolicy)) {
 		retval = PTR_ERR(p->mempolicy);
 		p->mempolicy = NULL;
1003
 		goto bad_fork_cleanup_cgroup;
Linus Torvalds's avatar
Linus Torvalds committed
1004
 	}
1005
	mpol_fix_fork_child_flag(p);
Linus Torvalds's avatar
Linus Torvalds committed
1006
#endif
1007 1008
#ifdef CONFIG_TRACE_IRQFLAGS
	p->irq_events = 0;
1009 1010 1011
#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
	p->hardirqs_enabled = 1;
#else
1012
	p->hardirqs_enabled = 0;
1013
#endif
1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025
	p->hardirq_enable_ip = 0;
	p->hardirq_enable_event = 0;
	p->hardirq_disable_ip = _THIS_IP_;
	p->hardirq_disable_event = 0;
	p->softirqs_enabled = 1;
	p->softirq_enable_ip = _THIS_IP_;
	p->softirq_enable_event = 0;
	p->softirq_disable_ip = 0;
	p->softirq_disable_event = 0;
	p->hardirq_context = 0;
	p->softirq_context = 0;
#endif
Ingo Molnar's avatar
Ingo Molnar committed
1026 1027 1028 1029 1030
#ifdef CONFIG_LOCKDEP
	p->lockdep_depth = 0; /* no locks held yet */
	p->curr_chain_key = 0;
	p->lockdep_recursion = 0;
#endif
Linus Torvalds's avatar
Linus Torvalds committed
1031

1032 1033 1034 1035
#ifdef CONFIG_DEBUG_MUTEXES
	p->blocked_on = NULL; /* not blocked yet */
#endif

1036 1037 1038
	/* Perform scheduler related setup. Assign this task to a CPU. */
	sched_fork(p, clone_flags);

Linus Torvalds's avatar
Linus Torvalds committed
1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057
	if ((retval = security_task_alloc(p)))
		goto bad_fork_cleanup_policy;
	if ((retval = audit_alloc(p)))
		goto bad_fork_cleanup_security;
	/* copy all the process information */
	if ((retval = copy_semundo(clone_flags, p)))
		goto bad_fork_cleanup_audit;
	if ((retval = copy_files(clone_flags, p)))
		goto bad_fork_cleanup_semundo;
	if ((retval = copy_fs(clone_flags, p)))
		goto bad_fork_cleanup_files;
	if ((retval = copy_sighand(clone_flags, p)))
		goto bad_fork_cleanup_fs;
	if ((retval = copy_signal(clone_flags, p)))
		goto bad_fork_cleanup_sighand;
	if ((retval = copy_mm(clone_flags, p)))
		goto bad_fork_cleanup_signal;
	if ((retval = copy_keys(clone_flags, p)))
		goto bad_fork_cleanup_mm;
1058
	if ((retval = copy_namespaces(clone_flags, p)))
Linus Torvalds's avatar
Linus Torvalds committed
1059
		goto bad_fork_cleanup_keys;
1060
	if ((retval = copy_io(clone_flags, p)))
1061
		goto bad_fork_cleanup_namespaces;
Linus Torvalds's avatar
Linus Torvalds committed
1062 1063
	retval = copy_thread(0, clone_flags, stack_start, stack_size, p, regs);
	if (retval)
1064
		goto bad_fork_cleanup_io;
Linus Torvalds's avatar
Linus Torvalds committed
1065

1066 1067 1068 1069
	if (pid != &init_struct_pid) {
		retval = -ENOMEM;
		pid = alloc_pid(task_active_pid_ns(p));
		if (!pid)
1070
			goto bad_fork_cleanup_io;
1071 1072 1073 1074 1075 1076

		if (clone_flags & CLONE_NEWPID) {
			retval = pid_ns_prepare_proc(task_active_pid_ns(p));
			if (retval < 0)
				goto bad_fork_free_pid;
		}
1077 1078 1079 1080 1081 1082 1083
	}

	p->pid = pid_nr(pid);
	p->tgid = p->pid;
	if (clone_flags & CLONE_THREAD)
		p->tgid = current->tgid;

Linus Torvalds's avatar
Linus Torvalds committed
1084 1085 1086 1087 1088
	p->set_child_tid = (clone_flags & CLONE_CHILD_SETTID) ? child_tidptr : NULL;
	/*
	 * Clear TID on mm_release()?
	 */
	p->clear_child_tid = (clone_flags & CLONE_CHILD_CLEARTID) ? child_tidptr: NULL;
1089
#ifdef CONFIG_FUTEX
1090 1091 1092 1093
	p->robust_list = NULL;
#ifdef CONFIG_COMPAT
	p->compat_robust_list = NULL;
#endif
1094 1095
	INIT_LIST_HEAD(&p->pi_state_list);
	p->pi_state_cache = NULL;
1096
#endif
1097 1098 1099 1100 1101 1102
	/*
	 * sigaltstack should be cleared when sharing the same VM
	 */
	if ((clone_flags & (CLONE_VM|CLONE_VFORK)) == CLONE_VM)
		p->sas_ss_sp = p->sas_ss_size = 0;

Linus Torvalds's avatar
Linus Torvalds committed
1103 1104 1105 1106 1107
	/*
	 * Syscall tracing should be turned off in the child regardless
	 * of CLONE_PTRACE.
	 */
	clear_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
1108 1109 1110
#ifdef TIF_SYSCALL_EMU
	clear_tsk_thread_flag(p, TIF_SYSCALL_EMU);
#endif
Arjan van de Ven's avatar
Arjan van de Ven committed
1111
	clear_all_latency_tracing(p);
Linus Torvalds's avatar
Linus Torvalds committed
1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126

	/* Our parent execution domain becomes current domain
	   These must match for thread signalling to apply */
	p->parent_exec_id = p->self_exec_id;

	/* ok, now we should be set up.. */
	p->exit_signal = (clone_flags & CLONE_THREAD) ? -1 : (clone_flags & CSIGNAL);
	p->pdeath_signal = 0;
	p->exit_state = 0;

	/*
	 * Ok, make it visible to the rest of the system.
	 * We dont wake it up yet.
	 */
	p->group_leader = p;
1127
	INIT_LIST_HEAD(&p->thread_group);
Linus Torvalds's avatar
Linus Torvalds committed
1128 1129 1130
	INIT_LIST_HEAD(&p->ptrace_children);
	INIT_LIST_HEAD(&p->ptrace_list);

1131 1132 1133 1134 1135 1136
	/* Now that the task is set up, run cgroup callbacks if
	 * necessary. We need to run them before the task is visible
	 * on the tasklist. */
	cgroup_fork_callbacks(p);
	cgroup_callbacks_done = 1;

Linus Torvalds's avatar
Linus Torvalds committed
1137 1138 1139 1140
	/* Need tasklist lock for parent etc handling! */
	write_lock_irq(&tasklist_lock);

	/*
1141 1142 1143 1144 1145 1146 1147
	 * The task hasn't been attached yet, so its cpus_allowed mask will
	 * not be changed, nor will its assigned CPU.
	 *
	 * The cpus_allowed mask of the parent may have changed after it was
	 * copied first time - so re-copy it here, then check the child's CPU
	 * to ensure it is on a valid CPU (and if not, just force it back to
	 * parent's CPU). This avoids alot of nasty races.
Linus Torvalds's avatar
Linus Torvalds committed
1148 1149
	 */
	p->cpus_allowed = current->cpus_allowed;
Peter Zijlstra's avatar
Peter Zijlstra committed
1150
	p->rt.nr_cpus_allowed = current->rt.nr_cpus_allowed;
1151 1152
	if (unlikely(!cpu_isset(task_cpu(p), p->cpus_allowed) ||
			!cpu_online(task_cpu(p))))
1153
		set_task_cpu(p, smp_processor_id());
Linus Torvalds's avatar
Linus Torvalds committed
1154 1155 1156 1157 1158 1159 1160 1161

	/* CLONE_PARENT re-uses the old parent */
	if (clone_flags & (CLONE_PARENT|CLONE_THREAD))
		p->real_parent = current->real_parent;
	else
		p->real_parent = current;
	p->parent = p->real_parent;

1162
	spin_lock(&current->sighand->siglock);
1163 1164 1165 1166 1167 1168 1169 1170 1171

	/*
	 * Process group and session signals need to be delivered to just the
	 * parent before the fork or both the parent and the child after the
	 * fork. Restart if a signal comes in before we add the new process to
	 * it's process group.
	 * A fatal signal pending means that current will exit, so the new
	 * thread can't slip out of an OOM kill (or normal SIGKILL).
 	 */
Daniel Walker's avatar
Daniel Walker committed
1172
	recalc_sigpending();
1173 1174 1175 1176
	if (signal_pending(current)) {
		spin_unlock(&current->sighand->siglock);
		write_unlock_irq(&tasklist_lock);
		retval = -ERESTARTNOINTR;
1177
		goto bad_fork_free_pid;
1178 1179
	}

Linus Torvalds's avatar
Linus Torvalds committed
1180 1181
	if (clone_flags & CLONE_THREAD) {
		p->group_leader = current->group_leader;
1182
		list_add_tail_rcu(&p->thread_group, &p->group_leader->thread_group);
Linus Torvalds's avatar
Linus Torvalds committed
1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199

		if (!cputime_eq(current->signal->it_virt_expires,
				cputime_zero) ||
		    !cputime_eq(current->signal->it_prof_expires,
				cputime_zero) ||
		    current->signal->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY ||
		    !list_empty(&current->signal->cpu_timers[0]) ||
		    !list_empty(&current->signal->cpu_timers[1]) ||
		    !list_empty(&current->signal->cpu_timers[2])) {
			/*
			 * Have child wake up on its first tick to check
			 * for process CPU timers.
			 */
			p->it_prof_expires = jiffies_to_cputime(1);
		}
	}

1200 1201 1202 1203 1204 1205
	if (likely(p->pid)) {
		add_parent(p);
		if (unlikely(p->ptrace & PT_PTRACED))
			__ptrace_link(p, current->parent);

		if (thread_group_leader(p)) {
Eric W. Biederman's avatar
Eric W. Biederman committed
1206
			if (clone_flags & CLONE_NEWPID)
1207
				p->nsproxy->pid_ns->child_reaper = p;
1208

1209
			p->signal->leader_pid = pid;
Eric W. Biederman's avatar
Eric W. Biederman committed
1210 1211 1212 1213 1214
			p->signal->tty = current->signal->tty;
			set_task_pgrp(p, task_pgrp_nr(current));
			set_task_session(p, task_session_nr(current));
			attach_pid(p, PIDTYPE_PGID, task_pgrp(current));
			attach_pid(p, PIDTYPE_SID, task_session(current));
1215
			list_add_tail_rcu(&p->tasks, &init_task.tasks);
Linus Torvalds's avatar
Linus Torvalds committed
1216
			__get_cpu_var(process_counts)++;
1217
		}
1218
		attach_pid(p, PIDTYPE_PID, pid);
1219
		nr_threads++;
Linus Torvalds's avatar
Linus Torvalds committed
1220 1221 1222
	}

	total_forks++;
1223
	spin_unlock(&current->sighand->siglock);
Linus Torvalds's avatar
Linus Torvalds committed
1224
	write_unlock_irq(&tasklist_lock);
1225
	proc_fork_connector(p);
1226
	cgroup_post_fork(p);
Linus Torvalds's avatar
Linus Torvalds committed
1227 1228
	return p;

1229 1230 1231
bad_fork_free_pid:
	if (pid != &init_struct_pid)
		free_pid(pid);
1232 1233
bad_fork_cleanup_io:
	put_io_context(p->io_context);
1234
bad_fork_cleanup_namespaces:
1235
	exit_task_namespaces(p);
Linus Torvalds's avatar
Linus Torvalds committed
1236 1237 1238 1239 1240 1241
bad_fork_cleanup_keys:
	exit_keys(p);
bad_fork_cleanup_mm:
	if (p->mm)
		mmput(p->mm);
bad_fork_cleanup_signal:
1242
	cleanup_signal(p);
Linus Torvalds's avatar
Linus Torvalds committed
1243
bad_fork_cleanup_sighand:
1244
	__cleanup_sighand(p->sighand);
Linus Torvalds's avatar
Linus Torvalds committed
1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256
bad_fork_cleanup_fs:
	exit_fs(p); /* blocking */
bad_fork_cleanup_files:
	exit_files(p); /* blocking */
bad_fork_cleanup_semundo:
	exit_sem(p);
bad_fork_cleanup_audit:
	audit_free(p);
bad_fork_cleanup_security:
	security_task_free(p);
bad_fork_cleanup_policy:
#ifdef CONFIG_NUMA
1257
	mpol_put(p->mempolicy);
1258
bad_fork_cleanup_cgroup:
Linus Torvalds's avatar
Linus Torvalds committed
1259
#endif
1260
	cgroup_exit(p, cgroup_callbacks_done);
1261
	delayacct_tsk_free(p);
Linus Torvalds's avatar
Linus Torvalds committed
1262 1263 1264
	if (p->binfmt)
		module_put(p->binfmt->module);
bad_fork_cleanup_put_domain:
1265
	module_put(task_thread_info(p)->exec_domain->module);
Linus Torvalds's avatar
Linus Torvalds committed
1266 1267 1268 1269 1270 1271
bad_fork_cleanup_count:
	put_group_info(p->group_info);
	atomic_dec(&p->user->processes);
	free_uid(p->user);
bad_fork_free:
	free_task(p);
1272 1273
fork_out:
	return ERR_PTR(retval);
Linus Torvalds's avatar
Linus Torvalds committed
1274 1275
}

Adrian Bunk's avatar
Adrian Bunk committed
1276
noinline struct pt_regs * __cpuinit __attribute__((weak)) idle_regs(struct pt_regs *regs)
Linus Torvalds's avatar
Linus Torvalds committed
1277 1278 1279 1280 1281
{
	memset(regs, 0, sizeof(struct pt_regs));
	return regs;
}

1282
struct task_struct * __cpuinit fork_idle(int cpu)
Linus Torvalds's avatar
Linus Torvalds committed
1283
{
1284
	struct task_struct *task;
Linus Torvalds's avatar
Linus Torvalds committed
1285 1286
	struct pt_regs regs;

1287
	task = copy_process(CLONE_VM, 0, idle_regs(&regs), 0, NULL,
1288
				&init_struct_pid);
1289 1290
	if (!IS_ERR(task))
		init_idle(task, cpu);
1291

Linus Torvalds's avatar
Linus Torvalds committed
1292 1293 1294
	return task;
}

Alexey Dobriyan's avatar
Alexey Dobriyan committed
1295
static int fork_traceflag(unsigned clone_flags)
Linus Torvalds's avatar
Linus Torvalds committed
1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325
{
	if (clone_flags & CLONE_UNTRACED)
		return 0;
	else if (clone_flags & CLONE_VFORK) {
		if (current->ptrace & PT_TRACE_VFORK)
			return PTRACE_EVENT_VFORK;
	} else if ((clone_flags & CSIGNAL) != SIGCHLD) {
		if (current->ptrace & PT_TRACE_CLONE)
			return PTRACE_EVENT_CLONE;
	} else if (current->ptrace & PT_TRACE_FORK)
		return PTRACE_EVENT_FORK;

	return 0;
}

/*
 *  Ok, this is the main fork-routine.
 *
 * It copies the process, and if successful kick-starts
 * it and waits for it to finish using the VM if required.
 */
long do_fork(unsigned long clone_flags,
	      unsigned long stack_start,
	      struct pt_regs *regs,
	      unsigned long stack_size,
	      int __user *parent_tidptr,
	      int __user *child_tidptr)
{
	struct task_struct *p;
	int trace = 0;
1326
	long nr;
Linus Torvalds's avatar
Linus Torvalds committed
1327

1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344
	/*
	 * We hope to recycle these flags after 2.6.26
	 */
	if (unlikely(clone_flags & CLONE_STOPPED)) {
		static int __read_mostly count = 100;

		if (count > 0 && printk_ratelimit()) {
			char comm[TASK_COMM_LEN];

			count--;
			printk(KERN_INFO "fork(): process `%s' used deprecated "
					"clone flags 0x%lx\n",
				get_task_comm(comm, current),
				clone_flags & CLONE_STOPPED);
		}
	}

Linus Torvalds's avatar
Linus Torvalds committed
1345 1346 1347 1348 1349 1350
	if (unlikely(current->ptrace)) {
		trace = fork_traceflag (clone_flags);
		if (trace)
			clone_flags |= CLONE_PTRACE;
	}

1351
	p = copy_process(clone_flags, stack_start, regs, stack_size,
1352
			child_tidptr, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
1353 1354 1355 1356 1357 1358 1359
	/*
	 * Do this prior waking up the new thread - the thread pointer
	 * might get invalid after that point, if the thread exits quickly.
	 */
	if (!IS_ERR(p)) {
		struct completion vfork;

1360
		nr = task_pid_vnr(p);
1361 1362 1363

		if (clone_flags & CLONE_PARENT_SETTID)
			put_user(nr, parent_tidptr);
1364

Linus Torvalds's avatar
Linus Torvalds committed
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380
		if (clone_flags & CLONE_VFORK) {
			p->vfork_done = &vfork;
			init_completion(&vfork);
		}

		if ((p->ptrace & PT_PTRACED) || (clone_flags & CLONE_STOPPED)) {
			/*
			 * We'll start up with an immediate SIGSTOP.
			 */
			sigaddset(&p->pending.signal, SIGSTOP);
			set_tsk_thread_flag(p, TIF_SIGPENDING);
		}

		if (!(clone_flags & CLONE_STOPPED))
			wake_up_new_task(p, clone_flags);
		else
1381
			__set_task_state(p, TASK_STOPPED);
Linus Torvalds's avatar
Linus Torvalds committed
1382 1383

		if (unlikely (trace)) {
1384
			current->ptrace_message = nr;
Linus Torvalds's avatar
Linus Torvalds committed
1385 1386 1387 1388
			ptrace_notify ((trace << 8) | SIGTRAP);
		}

		if (clone_flags & CLONE_VFORK) {
1389
			freezer_do_not_count();
Linus Torvalds's avatar
Linus Torvalds committed
1390
			wait_for_completion(&vfork);
1391
			freezer_count();
1392 1393
			if (unlikely (current->ptrace & PT_TRACE_VFORK_DONE)) {
				current->ptrace_message = nr;
Linus Torvalds's avatar
Linus Torvalds committed
1394
				ptrace_notify ((PTRACE_EVENT_VFORK_DONE << 8) | SIGTRAP);
1395
			}
Linus Torvalds's avatar
Linus Torvalds committed
1396 1397
		}
	} else {
1398
		nr = PTR_ERR(p);
Linus Torvalds's avatar
Linus Torvalds committed
1399
	}
1400
	return nr;
Linus Torvalds's avatar
Linus Torvalds committed
1401 1402
}

1403 1404 1405 1406
#ifndef ARCH_MIN_MMSTRUCT_ALIGN
#define ARCH_MIN_MMSTRUCT_ALIGN 0
#endif

1407
static void sighand_ctor(struct kmem_cache *cachep, void *data)
1408 1409 1410
{
	struct sighand_struct *sighand = data;

1411
	spin_lock_init(&sighand->siglock);
Davide Libenzi's avatar
Davide Libenzi committed
1412
	init_waitqueue_head(&sighand->signalfd_wqh);
1413 1414
}

Linus Torvalds's avatar
Linus Torvalds committed
1415 1416 1417 1418
void __init proc_caches_init(void)
{
	sighand_cachep = kmem_cache_create("sighand_cache",
			sizeof(struct sighand_struct), 0,
1419
			SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_DESTROY_BY_RCU,
1420
			sighand_ctor);
Linus Torvalds's avatar
Linus Torvalds committed
1421 1422
	signal_cachep = kmem_cache_create("signal_cache",
			sizeof(struct signal_struct), 0,
1423 1424
			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
	files_cachep = kmem_cache_create("files_cache",
Linus Torvalds's avatar
Linus Torvalds committed
1425
			sizeof(struct files_struct), 0,
1426 1427
			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
	fs_cachep = kmem_cache_create("fs_cache",
Linus Torvalds's avatar
Linus Torvalds committed
1428
			sizeof(struct fs_struct), 0,
1429
			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
1430 1431
	vm_area_cachep = kmem_cache_create("vm_area_struct",
			sizeof(struct vm_area_struct), 0,
1432
			SLAB_PANIC, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
1433
	mm_cachep = kmem_cache_create("mm_struct",
1434
			sizeof(struct mm_struct), ARCH_MIN_MMSTRUCT_ALIGN,
1435
			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
1436
}
1437 1438 1439 1440 1441

/*
 * Check constraints on flags passed to the unshare system call and
 * force unsharing of additional process context as appropriate.
 */
Alexey Dobriyan's avatar
Alexey Dobriyan committed
1442
static void check_unshare_flags(unsigned long *flags_ptr)
1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483
{
	/*
	 * If unsharing a thread from a thread group, must also
	 * unshare vm.
	 */
	if (*flags_ptr & CLONE_THREAD)
		*flags_ptr |= CLONE_VM;

	/*
	 * If unsharing vm, must also unshare signal handlers.
	 */
	if (*flags_ptr & CLONE_VM)
		*flags_ptr |= CLONE_SIGHAND;

	/*
	 * If unsharing signal handlers and the task was created
	 * using CLONE_THREAD, then must unshare the thread
	 */
	if ((*flags_ptr & CLONE_SIGHAND) &&
	    (atomic_read(&current->signal->count) > 1))
		*flags_ptr |= CLONE_THREAD;

	/*
	 * If unsharing namespace, must also unshare filesystem information.
	 */
	if (*flags_ptr & CLONE_NEWNS)
		*flags_ptr |= CLONE_FS;
}

/*
 * Unsharing of tasks created with CLONE_THREAD is not supported yet
 */
static int unshare_thread(unsigned long unshare_flags)
{
	if (unshare_flags & CLONE_THREAD)
		return -EINVAL;

	return 0;
}

/*
1484
 * Unshare the filesystem structure if it is being shared
1485 1486 1487 1488 1489 1490
 */
static int unshare_fs(unsigned long unshare_flags, struct fs_struct **new_fsp)
{
	struct fs_struct *fs = current->fs;

	if ((unshare_flags & CLONE_FS) &&
1491 1492 1493 1494 1495
	    (fs && atomic_read(&fs->count) > 1)) {
		*new_fsp = __copy_fs_struct(current->fs);
		if (!*new_fsp)
			return -ENOMEM;
	}
1496 1497 1498 1499 1500

	return 0;
}

/*
1501
 * Unsharing of sighand is not supported yet
1502 1503 1504 1505 1506
 */
static int unshare_sighand(unsigned long unshare_flags, struct sighand_struct **new_sighp)
{
	struct sighand_struct *sigh = current->sighand;

1507
	if ((unshare_flags & CLONE_SIGHAND) && atomic_read(&sigh->count) > 1)
1508 1509 1510 1511 1512 1513
		return -EINVAL;
	else
		return 0;
}

/*
1514
 * Unshare vm if it is being shared
1515 1516 1517 1518 1519 1520
 */
static int unshare_vm(unsigned long unshare_flags, struct mm_struct **new_mmp)
{
	struct mm_struct *mm = current->mm;

	if ((unshare_flags & CLONE_VM) &&