fork.c 41.1 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
#include <linux/security.h>
36
#include <linux/hugetlb.h>
Linus Torvalds's avatar
Linus Torvalds committed
37 38 39
#include <linux/swap.h>
#include <linux/syscalls.h>
#include <linux/jiffies.h>
Roland McGrath's avatar
Roland McGrath committed
40
#include <linux/tracehook.h>
Linus Torvalds's avatar
Linus Torvalds committed
41
#include <linux/futex.h>
42
#include <linux/task_io_accounting_ops.h>
43
#include <linux/rcupdate.h>
Linus Torvalds's avatar
Linus Torvalds committed
44 45 46
#include <linux/ptrace.h>
#include <linux/mount.h>
#include <linux/audit.h>
47
#include <linux/memcontrol.h>
Linus Torvalds's avatar
Linus Torvalds committed
48 49 50
#include <linux/profile.h>
#include <linux/rmap.h>
#include <linux/acct.h>
51
#include <linux/tsacct_kern.h>
52
#include <linux/cn_proc.h>
Rafael J. Wysocki's avatar
Rafael J. Wysocki committed
53
#include <linux/freezer.h>
54
#include <linux/delayacct.h>
55
#include <linux/taskstats_kern.h>
56
#include <linux/random.h>
Miloslav Trmac's avatar
Miloslav Trmac committed
57
#include <linux/tty.h>
58
#include <linux/proc_fs.h>
59
#include <linux/blkdev.h>
Linus Torvalds's avatar
Linus Torvalds committed
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77

#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;

78
__cacheline_aligned DEFINE_RWLOCK(tasklist_lock);  /* outer */
Linus Torvalds's avatar
Linus Torvalds committed
79 80 81 82 83 84 85 86 87 88 89 90 91 92 93

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))
94
static struct kmem_cache *task_struct_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
95 96
#endif

97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113
#ifndef __HAVE_ARCH_THREAD_INFO_ALLOCATOR
static inline struct thread_info *alloc_thread_info(struct task_struct *tsk)
{
#ifdef CONFIG_DEBUG_STACK_USAGE
	gfp_t mask = GFP_KERNEL | __GFP_ZERO;
#else
	gfp_t mask = GFP_KERNEL;
#endif
	return (struct thread_info *)__get_free_pages(mask, THREAD_SIZE_ORDER);
}

static inline void free_thread_info(struct thread_info *ti)
{
	free_pages((unsigned long)ti, THREAD_SIZE_ORDER);
}
#endif

Linus Torvalds's avatar
Linus Torvalds committed
114
/* SLAB cache for signal_struct structures (tsk->signal) */
115
static struct kmem_cache *signal_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
116 117

/* SLAB cache for sighand_struct structures (tsk->sighand) */
118
struct kmem_cache *sighand_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
119 120

/* SLAB cache for files_struct structures (tsk->files) */
121
struct kmem_cache *files_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
122 123

/* SLAB cache for fs_struct structures (tsk->fs) */
124
struct kmem_cache *fs_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
125 126

/* SLAB cache for vm_area_struct structures */
127
struct kmem_cache *vm_area_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
128 129

/* SLAB cache for mm_struct structures (tsk->mm) */
130
static struct kmem_cache *mm_cachep;
Linus Torvalds's avatar
Linus Torvalds committed
131 132 133

void free_task(struct task_struct *tsk)
{
Peter Zijlstra's avatar
Peter Zijlstra committed
134
	prop_local_destroy_single(&tsk->dirties);
Roman Zippel's avatar
Roman Zippel committed
135
	free_thread_info(tsk->stack);
Ingo Molnar's avatar
Ingo Molnar committed
136
	rt_mutex_debug_task_free(tsk);
Linus Torvalds's avatar
Linus Torvalds committed
137 138 139 140
	free_task_struct(tsk);
}
EXPORT_SYMBOL(free_task);

141
void __put_task_struct(struct task_struct *tsk)
Linus Torvalds's avatar
Linus Torvalds committed
142
{
Eugene Teo's avatar
Eugene Teo committed
143
	WARN_ON(!tsk->exit_state);
Linus Torvalds's avatar
Linus Torvalds committed
144 145 146 147 148 149
	WARN_ON(atomic_read(&tsk->usage));
	WARN_ON(tsk == current);

	security_task_free(tsk);
	free_uid(tsk->user);
	put_group_info(tsk->group_info);
150
	delayacct_tsk_free(tsk);
Linus Torvalds's avatar
Linus Torvalds committed
151 152 153 154 155

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

Suresh Siddha's avatar
Suresh Siddha committed
156 157 158 159 160 161 162
/*
 * 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
163

Linus Torvalds's avatar
Linus Torvalds committed
164 165 166 167 168 169 170 171 172
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),
173
			ARCH_MIN_TASKALIGN, SLAB_PANIC, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
174 175
#endif

176 177 178
	/* do the arch specific task caches init */
	arch_task_cache_init();

Linus Torvalds's avatar
Linus Torvalds committed
179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197
	/*
	 * 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];
}

198 199 200 201 202 203 204
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
205 206 207 208
static struct task_struct *dup_task_struct(struct task_struct *orig)
{
	struct task_struct *tsk;
	struct thread_info *ti;
Peter Zijlstra's avatar
Peter Zijlstra committed
209
	int err;
Linus Torvalds's avatar
Linus Torvalds committed
210 211 212 213 214 215 216 217 218 219 220 221 222

	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;
	}

223 224 225 226
 	err = arch_dup_task_struct(tsk, orig);
	if (err)
		goto out;

Roman Zippel's avatar
Roman Zippel committed
227
	tsk->stack = ti;
Peter Zijlstra's avatar
Peter Zijlstra committed
228 229

	err = prop_local_init_single(&tsk->dirties);
230 231
	if (err)
		goto out;
Peter Zijlstra's avatar
Peter Zijlstra committed
232

233
	setup_thread_stack(tsk, orig);
Linus Torvalds's avatar
Linus Torvalds committed
234

235 236 237 238
#ifdef CONFIG_CC_STACKPROTECTOR
	tsk->stack_canary = get_random_int();
#endif

Linus Torvalds's avatar
Linus Torvalds committed
239 240
	/* One for us, one for whoever does the "release_task()" (usually parent) */
	atomic_set(&tsk->usage,2);
241
	atomic_set(&tsk->fs_excl, 0);
242
#ifdef CONFIG_BLK_DEV_IO_TRACE
243
	tsk->btrace_seq = 0;
244
#endif
245
	tsk->splice_pipe = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
246
	return tsk;
247 248 249 250 251

out:
	free_thread_info(ti);
	free_task_struct(tsk);
	return NULL;
Linus Torvalds's avatar
Linus Torvalds committed
252 253 254
}

#ifdef CONFIG_MMU
Alexey Dobriyan's avatar
Alexey Dobriyan committed
255
static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm)
Linus Torvalds's avatar
Linus Torvalds committed
256
{
257
	struct vm_area_struct *mpnt, *tmp, **pprev;
Linus Torvalds's avatar
Linus Torvalds committed
258 259 260 261 262 263
	struct rb_node **rb_link, *rb_parent;
	int retval;
	unsigned long charge;
	struct mempolicy *pol;

	down_write(&oldmm->mmap_sem);
264
	flush_cache_dup_mm(oldmm);
265 266 267 268
	/*
	 * Not linked in yet - no deadlock potential:
	 */
	down_write_nested(&mm->mmap_sem, SINGLE_DEPTH_NESTING);
269

Linus Torvalds's avatar
Linus Torvalds committed
270 271 272 273
	mm->locked_vm = 0;
	mm->mmap = NULL;
	mm->mmap_cache = NULL;
	mm->free_area_cache = oldmm->mmap_base;
274
	mm->cached_hole_size = ~0UL;
Linus Torvalds's avatar
Linus Torvalds committed
275 276 277 278 279 280 281
	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;

282
	for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) {
Linus Torvalds's avatar
Linus Torvalds committed
283 284 285
		struct file *file;

		if (mpnt->vm_flags & VM_DONTCOPY) {
286 287
			long pages = vma_pages(mpnt);
			mm->total_vm -= pages;
288
			vm_stat_account(mm, mpnt->vm_flags, mpnt->vm_file,
289
								-pages);
Linus Torvalds's avatar
Linus Torvalds committed
290 291 292 293 294 295 296 297 298
			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;
		}
299
		tmp = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL);
Linus Torvalds's avatar
Linus Torvalds committed
300 301 302
		if (!tmp)
			goto fail_nomem;
		*tmp = *mpnt;
303
		pol = mpol_dup(vma_policy(mpnt));
Linus Torvalds's avatar
Linus Torvalds committed
304 305 306 307 308 309 310 311 312 313
		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) {
314
			struct inode *inode = file->f_path.dentry->d_inode;
Linus Torvalds's avatar
Linus Torvalds committed
315 316 317
			get_file(file);
			if (tmp->vm_flags & VM_DENYWRITE)
				atomic_dec(&inode->i_writecount);
Daniel Walker's avatar
Daniel Walker committed
318

Linus Torvalds's avatar
Linus Torvalds committed
319 320 321 322 323 324 325 326 327
			/* 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);
		}

328 329 330 331 332 333 334 335
		/*
		 * Clear hugetlb-related page reserves for children. This only
		 * affects MAP_PRIVATE mappings. Faults generated by the child
		 * are not guaranteed to succeed, even if read-only
		 */
		if (is_vm_hugetlb_page(tmp))
			reset_vma_resv_huge_pages(tmp);

Linus Torvalds's avatar
Linus Torvalds committed
336
		/*
337
		 * Link in the new vma and copy the page table entries.
Linus Torvalds's avatar
Linus Torvalds committed
338 339 340 341 342 343 344 345 346
		 */
		*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++;
347
		retval = copy_page_range(mm, oldmm, mpnt);
Linus Torvalds's avatar
Linus Torvalds committed
348 349 350 351 352 353 354

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

		if (retval)
			goto out;
	}
355 356
	/* a new mm has just been created */
	arch_dup_mmap(oldmm, mm);
Linus Torvalds's avatar
Linus Torvalds committed
357 358
	retval = 0;
out:
359
	up_write(&mm->mmap_sem);
360
	flush_tlb_mm(oldmm);
Linus Torvalds's avatar
Linus Torvalds committed
361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380
	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)
{
381
	pgd_free(mm, mm->pgd);
Linus Torvalds's avatar
Linus Torvalds committed
382 383 384 385 386 387 388
}
#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
389
__cacheline_aligned_in_smp DEFINE_SPINLOCK(mmlist_lock);
Linus Torvalds's avatar
Linus Torvalds committed
390

391
#define allocate_mm()	(kmem_cache_alloc(mm_cachep, GFP_KERNEL))
Linus Torvalds's avatar
Linus Torvalds committed
392 393 394 395
#define free_mm(mm)	(kmem_cache_free(mm_cachep, (mm)))

#include <linux/init_task.h>

396
static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p)
Linus Torvalds's avatar
Linus Torvalds committed
397 398 399 400 401
{
	atomic_set(&mm->mm_users, 1);
	atomic_set(&mm->mm_count, 1);
	init_rwsem(&mm->mmap_sem);
	INIT_LIST_HEAD(&mm->mmlist);
402 403
	mm->flags = (current->mm) ? current->mm->flags
				  : MMF_DUMP_FILTER_DEFAULT;
404
	mm->core_state = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
405
	mm->nr_ptes = 0;
406
	set_mm_counter(mm, file_rss, 0);
407
	set_mm_counter(mm, anon_rss, 0);
Linus Torvalds's avatar
Linus Torvalds committed
408 409 410 411
	spin_lock_init(&mm->page_table_lock);
	rwlock_init(&mm->ioctx_list_lock);
	mm->ioctx_list = NULL;
	mm->free_area_cache = TASK_UNMAPPED_BASE;
412
	mm->cached_hole_size = ~0UL;
413
	mm_init_owner(mm, p);
Linus Torvalds's avatar
Linus Torvalds committed
414 415 416 417 418

	if (likely(!mm_alloc_pgd(mm))) {
		mm->def_flags = 0;
		return mm;
	}
419

Linus Torvalds's avatar
Linus Torvalds committed
420 421 422 423 424 425 426 427 428 429 430 431 432 433
	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));
434
		mm = mm_init(mm, current);
Linus Torvalds's avatar
Linus Torvalds committed
435 436 437 438 439 440 441 442 443
	}
	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.
 */
444
void __mmdrop(struct mm_struct *mm)
Linus Torvalds's avatar
Linus Torvalds committed
445 446 447 448 449 450
{
	BUG_ON(mm == &init_mm);
	mm_free_pgd(mm);
	destroy_context(mm);
	free_mm(mm);
}
451
EXPORT_SYMBOL_GPL(__mmdrop);
Linus Torvalds's avatar
Linus Torvalds committed
452 453 454 455 456 457

/*
 * Decrement the use count and release all resources for an mm.
 */
void mmput(struct mm_struct *mm)
{
Andrew Morton's avatar
Andrew Morton committed
458 459
	might_sleep();

Linus Torvalds's avatar
Linus Torvalds committed
460 461 462
	if (atomic_dec_and_test(&mm->mm_users)) {
		exit_aio(mm);
		exit_mmap(mm);
Matt Helsley's avatar
Matt Helsley committed
463
		set_mm_exe_file(mm, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
464 465 466 467 468 469 470 471 472 473 474 475 476 477
		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
 *
478
 * Returns %NULL if the task has no mm.  Checks PF_KTHREAD (meaning
Linus Torvalds's avatar
Linus Torvalds committed
479 480 481 482 483 484 485 486 487 488 489 490
 * 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) {
491
		if (task->flags & PF_KTHREAD)
Linus Torvalds's avatar
Linus Torvalds committed
492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525
			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);
	}
526 527 528 529 530 531 532 533 534 535

	/*
	 * 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
536 537 538 539 540 541 542 543 544 545 546 547
		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);
	}
}

548 549 550 551
/*
 * Allocate a new mm structure and copy contents from the
 * mm structure of the passed in task structure.
 */
552
struct mm_struct *dup_mm(struct task_struct *tsk)
553 554 555 556 557 558 559 560 561 562 563 564 565
{
	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));

566 567 568 569
	/* Initializing for Swap token stuff */
	mm->token_priority = 0;
	mm->last_interval = 0;

570
	if (!mm_init(mm, tsk))
571 572 573 574 575
		goto fail_nomem;

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

Matt Helsley's avatar
Matt Helsley committed
576 577
	dup_mm_exe_file(oldmm, mm);

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
	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
603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629
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;
630
	mm = dup_mm(tsk);
Linus Torvalds's avatar
Linus Torvalds committed
631 632 633 634
	if (!mm)
		goto fail_nomem;

good_mm:
635 636 637 638
	/* Initializing for Swap token stuff */
	mm->token_priority = 0;
	mm->last_interval = 0;

Linus Torvalds's avatar
Linus Torvalds committed
639 640 641 642 643 644 645 646
	tsk->mm = mm;
	tsk->active_mm = mm;
	return 0;

fail_nomem:
	return retval;
}

Alexey Dobriyan's avatar
Alexey Dobriyan committed
647
static struct fs_struct *__copy_fs_struct(struct fs_struct *old)
Linus Torvalds's avatar
Linus Torvalds committed
648 649 650 651 652 653 654 655
{
	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
656 657 658 659 660 661 662
		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
663
		} else {
Jan Blunck's avatar
Jan Blunck committed
664 665
			fs->altroot.mnt = NULL;
			fs->altroot.dentry = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
666 667 668 669 670 671 672 673 674 675 676 677 678
		}
		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
679
static int copy_fs(unsigned long clone_flags, struct task_struct *tsk)
Linus Torvalds's avatar
Linus Torvalds committed
680 681 682 683 684 685 686 687 688 689 690
{
	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;
}

691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717
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;
}

718
static int copy_io(unsigned long clone_flags, struct task_struct *tsk)
719 720 721 722 723 724
{
#ifdef CONFIG_BLOCK
	struct io_context *ioc = current->io_context;

	if (!ioc)
		return 0;
725 726 727 728 729 730 731 732
	/*
	 * 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)) {
733 734 735 736 737 738 739 740 741 742
		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
743
static int copy_sighand(unsigned long clone_flags, struct task_struct *tsk)
Linus Torvalds's avatar
Linus Torvalds committed
744 745 746 747 748 749 750 751
{
	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
752
	rcu_assign_pointer(tsk->sighand, sig);
Linus Torvalds's avatar
Linus Torvalds committed
753 754 755 756 757 758 759
	if (!sig)
		return -ENOMEM;
	atomic_set(&sig->count, 1);
	memcpy(sig->action, current->sighand->action, sizeof(sig->action));
	return 0;
}

760
void __cleanup_sighand(struct sighand_struct *sighand)
761 762 763 764 765
{
	if (atomic_dec_and_test(&sighand->count))
		kmem_cache_free(sighand_cachep, sighand);
}

Alexey Dobriyan's avatar
Alexey Dobriyan committed
766
static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
Linus Torvalds's avatar
Linus Torvalds committed
767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793
{
	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;
794
	sig->curr_target = tsk;
Linus Torvalds's avatar
Linus Torvalds committed
795 796 797
	init_sigpending(&sig->shared_pending);
	INIT_LIST_HEAD(&sig->posix_timers);

798
	hrtimer_init(&sig->real_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
799
	sig->it_real_incr.tv64 = 0;
Linus Torvalds's avatar
Linus Torvalds committed
800 801 802 803 804 805 806 807
	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 */
808
	sig->tty_old_pgrp = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
809 810

	sig->utime = sig->stime = sig->cutime = sig->cstime = cputime_zero;
811 812
	sig->gtime = cputime_zero;
	sig->cgtime = cputime_zero;
Linus Torvalds's avatar
Linus Torvalds committed
813 814
	sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0;
	sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0;
815
	sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0;
816 817 818 819 820 821
#ifdef CONFIG_TASK_XACCT
	sig->rchar = sig->wchar = sig->syscr = sig->syscw = 0;
#endif
#ifdef CONFIG_TASK_IO_ACCOUNTING
	memset(&sig->ioac, 0, sizeof(sig->ioac));
#endif
822
	sig->sum_sched_runtime = 0;
Linus Torvalds's avatar
Linus Torvalds committed
823 824 825
	INIT_LIST_HEAD(&sig->cpu_timers[0]);
	INIT_LIST_HEAD(&sig->cpu_timers[1]);
	INIT_LIST_HEAD(&sig->cpu_timers[2]);
826
	taskstats_tgid_init(sig);
Linus Torvalds's avatar
Linus Torvalds committed
827 828 829 830 831 832 833 834 835 836 837 838 839

	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);
	}
840
	acct_init_pacct(&sig->pacct);
Linus Torvalds's avatar
Linus Torvalds committed
841

Miloslav Trmac's avatar
Miloslav Trmac committed
842 843
	tty_audit_fork(sig);

Linus Torvalds's avatar
Linus Torvalds committed
844 845 846
	return 0;
}

847 848 849 850 851 852
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
853
static void cleanup_signal(struct task_struct *tsk)
854 855 856 857 858 859 860 861 862
{
	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
863
static void copy_flags(unsigned long clone_flags, struct task_struct *p)
Linus Torvalds's avatar
Linus Torvalds committed
864 865 866
{
	unsigned long new_flags = p->flags;

867
	new_flags &= ~PF_SUPERPRIV;
Linus Torvalds's avatar
Linus Torvalds committed
868
	new_flags |= PF_FORKNOEXEC;
Roland McGrath's avatar
Roland McGrath committed
869
	new_flags |= PF_STARTING;
Linus Torvalds's avatar
Linus Torvalds committed
870
	p->flags = new_flags;
871
	clear_freeze_flag(p);
Linus Torvalds's avatar
Linus Torvalds committed
872 873 874 875 876 877
}

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

878
	return task_pid_vnr(current);
Linus Torvalds's avatar
Linus Torvalds committed
879 880
}

Alexey Dobriyan's avatar
Alexey Dobriyan committed
881
static void rt_mutex_init_task(struct task_struct *p)
Ingo Molnar's avatar
Ingo Molnar committed
882 883
{
	spin_lock_init(&p->pi_lock);
884
#ifdef CONFIG_RT_MUTEXES
Ingo Molnar's avatar
Ingo Molnar committed
885 886 887 888 889
	plist_head_init(&p->pi_waiters, &p->pi_lock);
	p->pi_blocked_on = NULL;
#endif
}

890 891 892 893 894 895 896
#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
897 898 899 900 901 902 903 904
/*
 * 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.
 */
905 906 907 908 909
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,
Roland McGrath's avatar
Roland McGrath committed
910 911
					struct pid *pid,
					int trace)
Linus Torvalds's avatar
Linus Torvalds committed
912 913
{
	int retval;
914
	struct task_struct *p;
915
	int cgroup_callbacks_done = 0;
Linus Torvalds's avatar
Linus Torvalds committed
916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943

	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;

944 945
	rt_mutex_init_task(p);

946
#ifdef CONFIG_PROVE_LOCKING
947 948 949
	DEBUG_LOCKS_WARN_ON(!p->hardirqs_enabled);
	DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled);
#endif
Linus Torvalds's avatar
Linus Torvalds committed
950 951 952 953
	retval = -EAGAIN;
	if (atomic_read(&p->user->processes) >=
			p->signal->rlim[RLIMIT_NPROC].rlim_cur) {
		if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RESOURCE) &&
954
		    p->user != current->nsproxy->user_ns->root_user)
Linus Torvalds's avatar
Linus Torvalds committed
955 956 957 958 959 960 961 962 963 964 965 966 967 968 969
			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;

970
	if (!try_module_get(task_thread_info(p)->exec_domain->module))
Linus Torvalds's avatar
Linus Torvalds committed
971 972 973 974 975 976
		goto bad_fork_cleanup_count;

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

	p->did_exec = 0;
977
	delayacct_tsk_init(p);	/* Must remain after dup_task_struct() */
Linus Torvalds's avatar
Linus Torvalds committed
978 979 980
	copy_flags(clone_flags, p);
	INIT_LIST_HEAD(&p->children);
	INIT_LIST_HEAD(&p->sibling);
Paul E. McKenney's avatar
Paul E. McKenney committed
981 982 983 984
#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
985 986 987 988 989 990 991 992
	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;
993
	p->gtime = cputime_zero;
994 995
	p->utimescaled = cputime_zero;
	p->stimescaled = cputime_zero;
996
	p->prev_utime = cputime_zero;
997
	p->prev_stime = cputime_zero;
998

999 1000 1001 1002 1003
#ifdef CONFIG_DETECT_SOFTLOCKUP
	p->last_switch_count = 0;
	p->last_switch_timestamp = 0;
#endif

1004
#ifdef CONFIG_TASK_XACCT
Linus Torvalds's avatar
Linus Torvalds committed
1005 1006 1007 1008
	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 */
1009
#endif
1010
	task_io_accounting_init(p);
Linus Torvalds's avatar
Linus Torvalds committed
1011 1012
	acct_clear_integrals(p);

Daniel Walker's avatar
Daniel Walker committed
1013
	p->it_virt_expires = cputime_zero;
Linus Torvalds's avatar
Linus Torvalds committed
1014
	p->it_prof_expires = cputime_zero;
Daniel Walker's avatar
Daniel Walker committed
1015 1016 1017 1018
	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
1019 1020 1021

	p->lock_depth = -1;		/* -1 = no lock */
	do_posix_clock_monotonic_gettime(&p->start_time);
1022 1023
	p->real_start_time = p->start_time;
	monotonic_to_bootbased(&p->real_start_time);
1024
#ifdef CONFIG_SECURITY
Linus Torvalds's avatar
Linus Torvalds committed
1025
	p->security = NULL;
1026
#endif
1027
	p->cap_bset = current->cap_bset;
Linus Torvalds's avatar
Linus Torvalds committed
1028 1029
	p->io_context = NULL;
	p->audit_context = NULL;
1030
	cgroup_fork(p);
Linus Torvalds's avatar
Linus Torvalds committed
1031
#ifdef CONFIG_NUMA
1032
	p->mempolicy = mpol_dup(p->mempolicy);
Linus Torvalds's avatar
Linus Torvalds committed
1033 1034 1035
 	if (IS_ERR(p->mempolicy)) {
 		retval = PTR_ERR(p->mempolicy);
 		p->mempolicy = NULL;
1036
 		goto bad_fork_cleanup_cgroup;
Linus Torvalds's avatar
Linus Torvalds committed
1037
 	}
1038
	mpol_fix_fork_child_flag(p);
Linus Torvalds's avatar
Linus Torvalds committed
1039
#endif
1040 1041
#ifdef CONFIG_TRACE_IRQFLAGS
	p->irq_events = 0;
1042 1043 1044
#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
	p->hardirqs_enabled = 1;
#else
1045
	p->hardirqs_enabled = 0;
1046
#endif
1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058
	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
1059 1060 1061 1062 1063
#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
1064

1065 1066 1067 1068
#ifdef CONFIG_DEBUG_MUTEXES
	p->blocked_on = NULL; /* not blocked yet */
#endif

1069 1070 1071
	/* Perform scheduler related setup. Assign this task to a CPU. */
	sched_fork(p, clone_flags);

Linus Torvalds's avatar
Linus Torvalds committed
1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090
	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;
1091
	if ((retval = copy_namespaces(clone_flags, p)))
Linus Torvalds's avatar
Linus Torvalds committed
1092
		goto bad_fork_cleanup_keys;
1093
	if ((retval = copy_io(clone_flags, p)))
1094
		goto bad_fork_cleanup_namespaces;
Linus Torvalds's avatar
Linus Torvalds committed
1095 1096
	retval = copy_thread(0, clone_flags, stack_start, stack_size, p, regs);
	if (retval)
1097
		goto bad_fork_cleanup_io;
Linus Torvalds's avatar
Linus Torvalds committed
1098

1099 1100 1101 1102
	if (pid != &init_struct_pid) {
		retval = -ENOMEM;
		pid = alloc_pid(task_active_pid_ns(p));
		if (!pid)
1103
			goto bad_fork_cleanup_io;
1104 1105 1106 1107 1108 1109

		if (clone_flags & CLONE_NEWPID) {
			retval = pid_ns_prepare_proc(task_active_pid_ns(p));
			if (retval < 0)
				goto bad_fork_free_pid;
		}
1110 1111 1112 1113 1114 1115 1116
	}

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

1117 1118 1119 1120 1121 1122
	if (current->nsproxy != p->nsproxy) {
		retval = ns_cgroup_clone(p, pid);
		if (retval)
			goto bad_fork_free_pid;
	}

Linus Torvalds's avatar
Linus Torvalds committed
1123 1124 1125 1126 1127
	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;
1128
#ifdef CONFIG_FUTEX
1129 1130 1131 1132
	p->robust_list = NULL;
#ifdef CONFIG_COMPAT
	p->compat_robust_list = NULL;
#endif
1133 1134
	INIT_LIST_HEAD(&p->pi_state_list);
	p->pi_state_cache = NULL;
1135
#endif
1136 1137 1138 1139 1140 1141
	/*
	 * 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
1142 1143 1144 1145 1146
	/*
	 * Syscall tracing should be turned off in the child regardless
	 * of CLONE_PTRACE.
	 */
	clear_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
1147 1148 1149
#ifdef TIF_SYSCALL_EMU
	clear_tsk_thread_flag(p, TIF_SYSCALL_EMU);
#endif
Arjan van de Ven's avatar
Arjan van de Ven committed
1150
	clear_all_latency_tracing(p);
Linus Torvalds's avatar
Linus Torvalds committed
1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165

	/* 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;
1166
	INIT_LIST_HEAD(&p->thread_group);
Linus Torvalds's avatar
Linus Torvalds committed
1167

1168 1169 1170 1171 1172 1173
	/* 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
1174 1175 1176 1177
	/* Need tasklist lock for parent etc handling! */
	write_lock_irq(&tasklist_lock);

	/*
1178 1179 1180 1181 1182 1183 1184
	 * 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
1185 1186
	 */
	p->cpus_allowed = current->cpus_allowed;
Peter Zijlstra's avatar
Peter Zijlstra committed
1187
	p->rt.nr_cpus_allowed = current->rt.nr_cpus_allowed;
1188 1189
	if (unlikely(!cpu_isset(task_cpu(p), p->cpus_allowed) ||
			!cpu_online(task_cpu(p))))
1190
		set_task_cpu(p, smp_processor_id());
Linus Torvalds's avatar
Linus Torvalds committed
1191 1192 1193 1194 1195 1196 1197

	/* 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;

1198
	spin_lock(&current->sighand->siglock);
1199 1200 1201 1202 1203 1204 1205 1206 1207

	/*
	 * 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
1208
	recalc_sigpending();
1209 1210 1211 1212
	if (signal_pending(current)) {
		spin_unlock(&current->sighand->siglock);
		write_unlock_irq(&tasklist_lock);
		retval = -ERESTARTNOINTR;
1213
		goto bad_fork_free_pid;
1214 1215
	}

Linus Torvalds's avatar
Linus Torvalds committed
1216 1217
	if (clone_flags & CLONE_THREAD) {
		p->group_leader = current->group_leader;
1218
		list_add_tail_rcu(&p->thread_group, &p->group_leader->thread_group);
Linus Torvalds's avatar
Linus Torvalds committed
1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235

		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);
		}
	}