base.c 59.1 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13
/*
 *  linux/fs/proc/base.c
 *
 *  Copyright (C) 1991, 1992 Linus Torvalds
 *
 *  proc base directory handling functions
 *
 *  1999, Al Viro. Rewritten. Now it covers the whole per-process part.
 *  Instead of using magical inumbers to determine the kind of object
 *  we allocate and fill in-core inodes upon lookup. They don't even
 *  go into icache. We cache the reference to task_struct upon lookup too.
 *  Eventually it should become a filesystem in its own. We don't use the
 *  rest of procfs anymore.
M
Mauricio Lin 已提交
14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
 *
 *
 *  Changelog:
 *  17-Jan-2005
 *  Allan Bezerra
 *  Bruna Moreira <bruna.moreira@indt.org.br>
 *  Edjard Mota <edjard.mota@indt.org.br>
 *  Ilias Biris <ilias.biris@indt.org.br>
 *  Mauricio Lin <mauricio.lin@indt.org.br>
 *
 *  Embedded Linux Lab - 10LE Instituto Nokia de Tecnologia - INdT
 *
 *  A new process specific entry (smaps) included in /proc. It shows the
 *  size of rss for each memory area. The maps entry lacks information
 *  about physical memory size (rss) for each mapped file, i.e.,
 *  rss information for executables and library files.
 *  This additional information is useful for any tools that need to know
 *  about physical memory consumption for a process specific library.
 *
 *  Changelog:
 *  21-Feb-2005
 *  Embedded Linux Lab - 10LE Instituto Nokia de Tecnologia - INdT
 *  Pud inclusion in the page table walking.
 *
 *  ChangeLog:
 *  10-Mar-2005
 *  10LE Instituto Nokia de Tecnologia - INdT:
 *  A better way to walks through the page table as suggested by Hugh Dickins.
 *
 *  Simo Piiroinen <simo.piiroinen@nokia.com>:
 *  Smaps information related to shared, private, clean and dirty pages.
 *
 *  Paul Mundt <paul.mundt@nokia.com>:
 *  Overall revision about smaps.
L
Linus Torvalds 已提交
48 49 50 51 52 53 54 55 56
 */

#include <asm/uaccess.h>

#include <linux/errno.h>
#include <linux/time.h>
#include <linux/proc_fs.h>
#include <linux/stat.h>
#include <linux/init.h>
57
#include <linux/capability.h>
L
Linus Torvalds 已提交
58 59 60 61
#include <linux/file.h>
#include <linux/string.h>
#include <linux/seq_file.h>
#include <linux/namei.h>
62
#include <linux/mnt_namespace.h>
L
Linus Torvalds 已提交
63
#include <linux/mm.h>
64
#include <linux/rcupdate.h>
L
Linus Torvalds 已提交
65
#include <linux/kallsyms.h>
K
Kees Cook 已提交
66
#include <linux/module.h>
L
Linus Torvalds 已提交
67 68 69 70 71
#include <linux/mount.h>
#include <linux/security.h>
#include <linux/ptrace.h>
#include <linux/cpuset.h>
#include <linux/audit.h>
A
Al Viro 已提交
72
#include <linux/poll.h>
73
#include <linux/nsproxy.h>
74
#include <linux/oom.h>
L
Linus Torvalds 已提交
75 76
#include "internal.h"

77 78 79 80 81 82 83 84 85 86
/* NOTE:
 *	Implementing inode permission operations in /proc is almost
 *	certainly an error.  Permission checks need to happen during
 *	each system call not at open time.  The reason is that most of
 *	what we wish to check for permissions in /proc varies at runtime.
 *
 *	The classic example of a problem is opening file descriptors
 *	in /proc for a task before it execs a suid executable.
 */

L
Linus Torvalds 已提交
87

88
/* Worst case buffer size needed for holding an integer. */
A
Andrew Morton 已提交
89
#define PROC_NUMBUF 13
90

L
Linus Torvalds 已提交
91 92
struct pid_entry {
	char *name;
93
	int len;
L
Linus Torvalds 已提交
94
	mode_t mode;
95
	const struct inode_operations *iop;
96
	const struct file_operations *fop;
97
	union proc_op op;
L
Linus Torvalds 已提交
98 99
};

100
#define NOD(NAME, MODE, IOP, FOP, OP) {			\
101
	.name = (NAME),					\
102
	.len  = sizeof(NAME) - 1,			\
103 104 105 106 107 108
	.mode = MODE,					\
	.iop  = IOP,					\
	.fop  = FOP,					\
	.op   = OP,					\
}

109 110
#define DIR(NAME, MODE, OTYPE)							\
	NOD(NAME, (S_IFDIR|(MODE)),						\
111 112
		&proc_##OTYPE##_inode_operations, &proc_##OTYPE##_operations,	\
		{} )
113 114
#define LNK(NAME, OTYPE)					\
	NOD(NAME, (S_IFLNK|S_IRWXUGO),				\
115 116
		&proc_pid_link_inode_operations, NULL,		\
		{ .proc_get_link = &proc_##OTYPE##_link } )
117 118
#define REG(NAME, MODE, OTYPE)				\
	NOD(NAME, (S_IFREG|(MODE)), NULL,		\
119
		&proc_##OTYPE##_operations, {})
120 121
#define INF(NAME, MODE, OTYPE)				\
	NOD(NAME, (S_IFREG|(MODE)), 			\
122 123
		NULL, &proc_info_file_operations,	\
		{ .proc_read = &proc_##OTYPE } )
L
Linus Torvalds 已提交
124

K
Kees Cook 已提交
125 126 127
int maps_protect;
EXPORT_SYMBOL(maps_protect);

128
static struct fs_struct *get_fs_struct(struct task_struct *task)
L
Linus Torvalds 已提交
129 130
{
	struct fs_struct *fs;
131 132
	task_lock(task);
	fs = task->fs;
L
Linus Torvalds 已提交
133 134
	if(fs)
		atomic_inc(&fs->count);
135 136 137 138
	task_unlock(task);
	return fs;
}

139 140 141 142 143 144 145 146 147 148 149 150 151
static int get_nr_threads(struct task_struct *tsk)
{
	/* Must be called with the rcu_read_lock held */
	unsigned long flags;
	int count = 0;

	if (lock_task_sighand(tsk, &flags)) {
		count = atomic_read(&tsk->signal->count);
		unlock_task_sighand(tsk, &flags);
	}
	return count;
}

152 153
static int proc_cwd_link(struct inode *inode, struct dentry **dentry, struct vfsmount **mnt)
{
154 155
	struct task_struct *task = get_proc_task(inode);
	struct fs_struct *fs = NULL;
156
	int result = -ENOENT;
157 158 159 160 161

	if (task) {
		fs = get_fs_struct(task);
		put_task_struct(task);
	}
L
Linus Torvalds 已提交
162 163 164 165 166 167 168 169 170 171 172 173 174
	if (fs) {
		read_lock(&fs->lock);
		*mnt = mntget(fs->pwdmnt);
		*dentry = dget(fs->pwd);
		read_unlock(&fs->lock);
		result = 0;
		put_fs_struct(fs);
	}
	return result;
}

static int proc_root_link(struct inode *inode, struct dentry **dentry, struct vfsmount **mnt)
{
175 176
	struct task_struct *task = get_proc_task(inode);
	struct fs_struct *fs = NULL;
L
Linus Torvalds 已提交
177
	int result = -ENOENT;
178 179 180 181 182

	if (task) {
		fs = get_fs_struct(task);
		put_task_struct(task);
	}
L
Linus Torvalds 已提交
183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205
	if (fs) {
		read_lock(&fs->lock);
		*mnt = mntget(fs->rootmnt);
		*dentry = dget(fs->root);
		read_unlock(&fs->lock);
		result = 0;
		put_fs_struct(fs);
	}
	return result;
}

#define MAY_PTRACE(task) \
	(task == current || \
	(task->parent == current && \
	(task->ptrace & PT_PTRACED) && \
	 (task->state == TASK_STOPPED || task->state == TASK_TRACED) && \
	 security_ptrace(current,task) == 0))

static int proc_pid_environ(struct task_struct *task, char * buffer)
{
	int res = 0;
	struct mm_struct *mm = get_task_mm(task);
	if (mm) {
206 207 208 209 210 211 212
		unsigned int len;

		res = -ESRCH;
		if (!ptrace_may_attach(task))
			goto out;

		len  = mm->env_end - mm->env_start;
L
Linus Torvalds 已提交
213 214 215
		if (len > PAGE_SIZE)
			len = PAGE_SIZE;
		res = access_process_vm(task, mm->env_start, buffer, len, 0);
216
out:
L
Linus Torvalds 已提交
217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284
		mmput(mm);
	}
	return res;
}

static int proc_pid_cmdline(struct task_struct *task, char * buffer)
{
	int res = 0;
	unsigned int len;
	struct mm_struct *mm = get_task_mm(task);
	if (!mm)
		goto out;
	if (!mm->arg_end)
		goto out_mm;	/* Shh! No looking before we're done */

 	len = mm->arg_end - mm->arg_start;
 
	if (len > PAGE_SIZE)
		len = PAGE_SIZE;
 
	res = access_process_vm(task, mm->arg_start, buffer, len, 0);

	// If the nul at the end of args has been overwritten, then
	// assume application is using setproctitle(3).
	if (res > 0 && buffer[res-1] != '\0' && len < PAGE_SIZE) {
		len = strnlen(buffer, res);
		if (len < res) {
		    res = len;
		} else {
			len = mm->env_end - mm->env_start;
			if (len > PAGE_SIZE - res)
				len = PAGE_SIZE - res;
			res += access_process_vm(task, mm->env_start, buffer+res, len, 0);
			res = strnlen(buffer, res);
		}
	}
out_mm:
	mmput(mm);
out:
	return res;
}

static int proc_pid_auxv(struct task_struct *task, char *buffer)
{
	int res = 0;
	struct mm_struct *mm = get_task_mm(task);
	if (mm) {
		unsigned int nwords = 0;
		do
			nwords += 2;
		while (mm->saved_auxv[nwords - 2] != 0); /* AT_NULL */
		res = nwords * sizeof(mm->saved_auxv[0]);
		if (res > PAGE_SIZE)
			res = PAGE_SIZE;
		memcpy(buffer, mm->saved_auxv, res);
		mmput(mm);
	}
	return res;
}


#ifdef CONFIG_KALLSYMS
/*
 * Provides a wchan file via kallsyms in a proper one-value-per-file format.
 * Returns the resolved symbol.  If that fails, simply return the address.
 */
static int proc_pid_wchan(struct task_struct *task, char *buffer)
{
A
Alexey Dobriyan 已提交
285
	unsigned long wchan;
286
	char symname[KSYM_NAME_LEN];
L
Linus Torvalds 已提交
287 288 289

	wchan = get_wchan(task);

290 291 292 293
	if (lookup_symbol_name(wchan, symname) < 0)
		return sprintf(buffer, "%lu", wchan);
	else
		return sprintf(buffer, "%s", symname);
L
Linus Torvalds 已提交
294 295 296 297 298 299 300 301 302
}
#endif /* CONFIG_KALLSYMS */

#ifdef CONFIG_SCHEDSTATS
/*
 * Provides /proc/PID/schedstat
 */
static int proc_pid_schedstat(struct task_struct *task, char *buffer)
{
303
	return sprintf(buffer, "%llu %llu %lu\n",
L
Linus Torvalds 已提交
304 305 306 307 308 309 310 311 312 313 314 315 316 317
			task->sched_info.cpu_time,
			task->sched_info.run_delay,
			task->sched_info.pcnt);
}
#endif

/* The badness from the OOM killer */
unsigned long badness(struct task_struct *p, unsigned long uptime);
static int proc_oom_score(struct task_struct *task, char *buffer)
{
	unsigned long points;
	struct timespec uptime;

	do_posix_clock_monotonic_gettime(&uptime);
318
	read_lock(&tasklist_lock);
L
Linus Torvalds 已提交
319
	points = badness(task, uptime.tv_sec);
320
	read_unlock(&tasklist_lock);
L
Linus Torvalds 已提交
321 322 323 324 325 326 327 328
	return sprintf(buffer, "%lu\n", points);
}

/************************************************************************/
/*                       Here the fs part begins                        */
/************************************************************************/

/* permission checks */
329
static int proc_fd_access_allowed(struct inode *inode)
L
Linus Torvalds 已提交
330
{
331 332
	struct task_struct *task;
	int allowed = 0;
333 334 335
	/* Allow access to a task's file descriptors if it is us or we
	 * may use ptrace attach to the process and find out that
	 * information.
336 337
	 */
	task = get_proc_task(inode);
338 339
	if (task) {
		allowed = ptrace_may_attach(task);
340
		put_task_struct(task);
341
	}
342
	return allowed;
L
Linus Torvalds 已提交
343 344
}

345 346 347 348 349 350 351 352 353
static int proc_setattr(struct dentry *dentry, struct iattr *attr)
{
	int error;
	struct inode *inode = dentry->d_inode;

	if (attr->ia_valid & ATTR_MODE)
		return -EPERM;

	error = inode_change_ok(inode, attr);
354 355
	if (!error)
		error = inode_setattr(inode, attr);
356 357 358
	return error;
}

359
static const struct inode_operations proc_def_inode_operations = {
360 361 362
	.setattr	= proc_setattr,
};

L
Linus Torvalds 已提交
363
extern struct seq_operations mounts_op;
A
Al Viro 已提交
364 365 366 367 368
struct proc_mounts {
	struct seq_file m;
	int event;
};

L
Linus Torvalds 已提交
369 370
static int mounts_open(struct inode *inode, struct file *file)
{
371
	struct task_struct *task = get_proc_task(inode);
372
	struct mnt_namespace *ns = NULL;
A
Al Viro 已提交
373 374
	struct proc_mounts *p;
	int ret = -EINVAL;
L
Linus Torvalds 已提交
375

376 377
	if (task) {
		task_lock(task);
378 379 380 381 382
		if (task->nsproxy) {
			ns = task->nsproxy->mnt_ns;
			if (ns)
				get_mnt_ns(ns);
		}
383 384 385
		task_unlock(task);
		put_task_struct(task);
	}
A
Al Viro 已提交
386

387
	if (ns) {
A
Al Viro 已提交
388 389 390 391 392 393
		ret = -ENOMEM;
		p = kmalloc(sizeof(struct proc_mounts), GFP_KERNEL);
		if (p) {
			file->private_data = &p->m;
			ret = seq_open(file, &mounts_op);
			if (!ret) {
394 395
				p->m.private = ns;
				p->event = ns->event;
A
Al Viro 已提交
396 397 398
				return 0;
			}
			kfree(p);
L
Linus Torvalds 已提交
399
		}
400
		put_mnt_ns(ns);
L
Linus Torvalds 已提交
401 402 403 404 405 406 407
	}
	return ret;
}

static int mounts_release(struct inode *inode, struct file *file)
{
	struct seq_file *m = file->private_data;
408 409
	struct mnt_namespace *ns = m->private;
	put_mnt_ns(ns);
L
Linus Torvalds 已提交
410 411 412
	return seq_release(inode, file);
}

A
Al Viro 已提交
413 414 415
static unsigned mounts_poll(struct file *file, poll_table *wait)
{
	struct proc_mounts *p = file->private_data;
416
	struct mnt_namespace *ns = p->m.private;
A
Al Viro 已提交
417 418 419 420 421 422 423 424 425 426 427 428 429 430
	unsigned res = 0;

	poll_wait(file, &ns->poll, wait);

	spin_lock(&vfsmount_lock);
	if (p->event != ns->event) {
		p->event = ns->event;
		res = POLLERR;
	}
	spin_unlock(&vfsmount_lock);

	return res;
}

431
static const struct file_operations proc_mounts_operations = {
L
Linus Torvalds 已提交
432 433 434 435
	.open		= mounts_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
	.release	= mounts_release,
A
Al Viro 已提交
436
	.poll		= mounts_poll,
L
Linus Torvalds 已提交
437 438
};

439 440 441 442 443 444 445
extern struct seq_operations mountstats_op;
static int mountstats_open(struct inode *inode, struct file *file)
{
	int ret = seq_open(file, &mountstats_op);

	if (!ret) {
		struct seq_file *m = file->private_data;
446
		struct mnt_namespace *mnt_ns = NULL;
447 448 449 450
		struct task_struct *task = get_proc_task(inode);

		if (task) {
			task_lock(task);
451
			if (task->nsproxy)
452 453 454
				mnt_ns = task->nsproxy->mnt_ns;
			if (mnt_ns)
				get_mnt_ns(mnt_ns);
455 456 457
			task_unlock(task);
			put_task_struct(task);
		}
458

459 460
		if (mnt_ns)
			m->private = mnt_ns;
461 462 463 464 465 466 467 468
		else {
			seq_release(inode, file);
			ret = -EINVAL;
		}
	}
	return ret;
}

469
static const struct file_operations proc_mountstats_operations = {
470 471 472 473 474 475
	.open		= mountstats_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
	.release	= mounts_release,
};

L
Linus Torvalds 已提交
476 477 478 479 480
#define PROC_BLOCK_SIZE	(3*1024)		/* 4K page size but our output routines use some slack for overruns */

static ssize_t proc_info_read(struct file * file, char __user * buf,
			  size_t count, loff_t *ppos)
{
481
	struct inode * inode = file->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
482 483
	unsigned long page;
	ssize_t length;
484 485 486 487 488
	struct task_struct *task = get_proc_task(inode);

	length = -ESRCH;
	if (!task)
		goto out_no_task;
L
Linus Torvalds 已提交
489 490 491

	if (count > PROC_BLOCK_SIZE)
		count = PROC_BLOCK_SIZE;
492 493

	length = -ENOMEM;
L
Linus Torvalds 已提交
494
	if (!(page = __get_free_page(GFP_KERNEL)))
495
		goto out;
L
Linus Torvalds 已提交
496 497 498 499 500 501

	length = PROC_I(inode)->op.proc_read(task, (char*)page);

	if (length >= 0)
		length = simple_read_from_buffer(buf, count, ppos, (char *)page, length);
	free_page(page);
502 503 504
out:
	put_task_struct(task);
out_no_task:
L
Linus Torvalds 已提交
505 506 507
	return length;
}

508
static const struct file_operations proc_info_file_operations = {
L
Linus Torvalds 已提交
509 510 511 512 513 514 515 516 517 518 519 520
	.read		= proc_info_read,
};

static int mem_open(struct inode* inode, struct file* file)
{
	file->private_data = (void*)((long)current->self_exec_id);
	return 0;
}

static ssize_t mem_read(struct file * file, char __user * buf,
			size_t count, loff_t *ppos)
{
521
	struct task_struct *task = get_proc_task(file->f_path.dentry->d_inode);
L
Linus Torvalds 已提交
522 523 524 525 526
	char *page;
	unsigned long src = *ppos;
	int ret = -ESRCH;
	struct mm_struct *mm;

527 528 529
	if (!task)
		goto out_no_task;

530
	if (!MAY_PTRACE(task) || !ptrace_may_attach(task))
L
Linus Torvalds 已提交
531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555
		goto out;

	ret = -ENOMEM;
	page = (char *)__get_free_page(GFP_USER);
	if (!page)
		goto out;

	ret = 0;
 
	mm = get_task_mm(task);
	if (!mm)
		goto out_free;

	ret = -EIO;
 
	if (file->private_data != (void*)((long)current->self_exec_id))
		goto out_put;

	ret = 0;
 
	while (count > 0) {
		int this_len, retval;

		this_len = (count > PAGE_SIZE) ? PAGE_SIZE : count;
		retval = access_process_vm(task, src, page, this_len, 0);
556
		if (!retval || !MAY_PTRACE(task) || !ptrace_may_attach(task)) {
L
Linus Torvalds 已提交
557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578
			if (!ret)
				ret = -EIO;
			break;
		}

		if (copy_to_user(buf, page, retval)) {
			ret = -EFAULT;
			break;
		}
 
		ret += retval;
		src += retval;
		buf += retval;
		count -= retval;
	}
	*ppos = src;

out_put:
	mmput(mm);
out_free:
	free_page((unsigned long) page);
out:
579 580
	put_task_struct(task);
out_no_task:
L
Linus Torvalds 已提交
581 582 583 584 585 586 587
	return ret;
}

#define mem_write NULL

#ifndef mem_write
/* This is a security hazard */
588
static ssize_t mem_write(struct file * file, const char __user *buf,
L
Linus Torvalds 已提交
589 590
			 size_t count, loff_t *ppos)
{
591
	int copied;
L
Linus Torvalds 已提交
592
	char *page;
593
	struct task_struct *task = get_proc_task(file->f_path.dentry->d_inode);
L
Linus Torvalds 已提交
594 595
	unsigned long dst = *ppos;

596 597 598 599
	copied = -ESRCH;
	if (!task)
		goto out_no_task;

600
	if (!MAY_PTRACE(task) || !ptrace_may_attach(task))
601
		goto out;
L
Linus Torvalds 已提交
602

603
	copied = -ENOMEM;
L
Linus Torvalds 已提交
604 605
	page = (char *)__get_free_page(GFP_USER);
	if (!page)
606
		goto out;
L
Linus Torvalds 已提交
607

608
	copied = 0;
L
Linus Torvalds 已提交
609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629
	while (count > 0) {
		int this_len, retval;

		this_len = (count > PAGE_SIZE) ? PAGE_SIZE : count;
		if (copy_from_user(page, buf, this_len)) {
			copied = -EFAULT;
			break;
		}
		retval = access_process_vm(task, dst, page, this_len, 1);
		if (!retval) {
			if (!copied)
				copied = -EIO;
			break;
		}
		copied += retval;
		buf += retval;
		dst += retval;
		count -= retval;			
	}
	*ppos = dst;
	free_page((unsigned long) page);
630 631 632
out:
	put_task_struct(task);
out_no_task:
L
Linus Torvalds 已提交
633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652
	return copied;
}
#endif

static loff_t mem_lseek(struct file * file, loff_t offset, int orig)
{
	switch (orig) {
	case 0:
		file->f_pos = offset;
		break;
	case 1:
		file->f_pos += offset;
		break;
	default:
		return -EINVAL;
	}
	force_successful_syscall_return();
	return file->f_pos;
}

653
static const struct file_operations proc_mem_operations = {
L
Linus Torvalds 已提交
654 655 656 657 658 659 660 661 662
	.llseek		= mem_lseek,
	.read		= mem_read,
	.write		= mem_write,
	.open		= mem_open,
};

static ssize_t oom_adjust_read(struct file *file, char __user *buf,
				size_t count, loff_t *ppos)
{
663
	struct task_struct *task = get_proc_task(file->f_path.dentry->d_inode);
664
	char buffer[PROC_NUMBUF];
L
Linus Torvalds 已提交
665
	size_t len;
666
	int oom_adjust;
L
Linus Torvalds 已提交
667

668 669 670 671 672
	if (!task)
		return -ESRCH;
	oom_adjust = task->oomkilladj;
	put_task_struct(task);

673
	len = snprintf(buffer, sizeof(buffer), "%i\n", oom_adjust);
674 675

	return simple_read_from_buffer(buf, count, ppos, buffer, len);
L
Linus Torvalds 已提交
676 677 678 679 680
}

static ssize_t oom_adjust_write(struct file *file, const char __user *buf,
				size_t count, loff_t *ppos)
{
681
	struct task_struct *task;
682
	char buffer[PROC_NUMBUF], *end;
L
Linus Torvalds 已提交
683 684
	int oom_adjust;

685 686 687
	memset(buffer, 0, sizeof(buffer));
	if (count > sizeof(buffer) - 1)
		count = sizeof(buffer) - 1;
L
Linus Torvalds 已提交
688 689 690
	if (copy_from_user(buffer, buf, count))
		return -EFAULT;
	oom_adjust = simple_strtol(buffer, &end, 0);
691 692
	if ((oom_adjust < OOM_ADJUST_MIN || oom_adjust > OOM_ADJUST_MAX) &&
	     oom_adjust != OOM_DISABLE)
L
Linus Torvalds 已提交
693 694 695
		return -EINVAL;
	if (*end == '\n')
		end++;
696
	task = get_proc_task(file->f_path.dentry->d_inode);
697 698
	if (!task)
		return -ESRCH;
699 700 701 702
	if (oom_adjust < task->oomkilladj && !capable(CAP_SYS_RESOURCE)) {
		put_task_struct(task);
		return -EACCES;
	}
L
Linus Torvalds 已提交
703
	task->oomkilladj = oom_adjust;
704
	put_task_struct(task);
L
Linus Torvalds 已提交
705 706 707 708 709
	if (end - buffer == 0)
		return -EIO;
	return end - buffer;
}

710
static const struct file_operations proc_oom_adjust_operations = {
L
Linus Torvalds 已提交
711 712 713 714
	.read		= oom_adjust_read,
	.write		= oom_adjust_write,
};

715
#ifdef CONFIG_MMU
716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748
static ssize_t clear_refs_write(struct file *file, const char __user *buf,
				size_t count, loff_t *ppos)
{
	struct task_struct *task;
	char buffer[PROC_NUMBUF], *end;
	struct mm_struct *mm;

	memset(buffer, 0, sizeof(buffer));
	if (count > sizeof(buffer) - 1)
		count = sizeof(buffer) - 1;
	if (copy_from_user(buffer, buf, count))
		return -EFAULT;
	if (!simple_strtol(buffer, &end, 0))
		return -EINVAL;
	if (*end == '\n')
		end++;
	task = get_proc_task(file->f_path.dentry->d_inode);
	if (!task)
		return -ESRCH;
	mm = get_task_mm(task);
	if (mm) {
		clear_refs_smap(mm);
		mmput(mm);
	}
	put_task_struct(task);
	if (end - buffer == 0)
		return -EIO;
	return end - buffer;
}

static struct file_operations proc_clear_refs_operations = {
	.write		= clear_refs_write,
};
749
#endif
750

L
Linus Torvalds 已提交
751 752 753 754 755
#ifdef CONFIG_AUDITSYSCALL
#define TMPBUFLEN 21
static ssize_t proc_loginuid_read(struct file * file, char __user * buf,
				  size_t count, loff_t *ppos)
{
756
	struct inode * inode = file->f_path.dentry->d_inode;
757
	struct task_struct *task = get_proc_task(inode);
L
Linus Torvalds 已提交
758 759 760
	ssize_t length;
	char tmpbuf[TMPBUFLEN];

761 762
	if (!task)
		return -ESRCH;
L
Linus Torvalds 已提交
763 764
	length = scnprintf(tmpbuf, TMPBUFLEN, "%u",
				audit_get_loginuid(task->audit_context));
765
	put_task_struct(task);
L
Linus Torvalds 已提交
766 767 768 769 770 771
	return simple_read_from_buffer(buf, count, ppos, tmpbuf, length);
}

static ssize_t proc_loginuid_write(struct file * file, const char __user * buf,
				   size_t count, loff_t *ppos)
{
772
	struct inode * inode = file->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
773 774 775 776 777 778 779
	char *page, *tmp;
	ssize_t length;
	uid_t loginuid;

	if (!capable(CAP_AUDIT_CONTROL))
		return -EPERM;

780
	if (current != pid_task(proc_pid(inode), PIDTYPE_PID))
L
Linus Torvalds 已提交
781 782
		return -EPERM;

783 784
	if (count >= PAGE_SIZE)
		count = PAGE_SIZE - 1;
L
Linus Torvalds 已提交
785 786 787 788 789 790 791 792 793 794 795 796

	if (*ppos != 0) {
		/* No partial writes. */
		return -EINVAL;
	}
	page = (char*)__get_free_page(GFP_USER);
	if (!page)
		return -ENOMEM;
	length = -EFAULT;
	if (copy_from_user(page, buf, count))
		goto out_free_page;

797
	page[count] = '\0';
L
Linus Torvalds 已提交
798 799 800 801 802 803
	loginuid = simple_strtoul(page, &tmp, 10);
	if (tmp == page) {
		length = -EINVAL;
		goto out_free_page;

	}
804
	length = audit_set_loginuid(current, loginuid);
L
Linus Torvalds 已提交
805 806 807 808 809 810 811 812
	if (likely(length == 0))
		length = count;

out_free_page:
	free_page((unsigned long) page);
	return length;
}

813
static const struct file_operations proc_loginuid_operations = {
L
Linus Torvalds 已提交
814 815 816 817 818
	.read		= proc_loginuid_read,
	.write		= proc_loginuid_write,
};
#endif

819 820 821 822 823 824 825 826 827 828 829 830 831 832 833
#ifdef CONFIG_FAULT_INJECTION
static ssize_t proc_fault_inject_read(struct file * file, char __user * buf,
				      size_t count, loff_t *ppos)
{
	struct task_struct *task = get_proc_task(file->f_dentry->d_inode);
	char buffer[PROC_NUMBUF];
	size_t len;
	int make_it_fail;

	if (!task)
		return -ESRCH;
	make_it_fail = task->make_it_fail;
	put_task_struct(task);

	len = snprintf(buffer, sizeof(buffer), "%i\n", make_it_fail);
834 835

	return simple_read_from_buffer(buf, count, ppos, buffer, len);
836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864
}

static ssize_t proc_fault_inject_write(struct file * file,
			const char __user * buf, size_t count, loff_t *ppos)
{
	struct task_struct *task;
	char buffer[PROC_NUMBUF], *end;
	int make_it_fail;

	if (!capable(CAP_SYS_RESOURCE))
		return -EPERM;
	memset(buffer, 0, sizeof(buffer));
	if (count > sizeof(buffer) - 1)
		count = sizeof(buffer) - 1;
	if (copy_from_user(buffer, buf, count))
		return -EFAULT;
	make_it_fail = simple_strtol(buffer, &end, 0);
	if (*end == '\n')
		end++;
	task = get_proc_task(file->f_dentry->d_inode);
	if (!task)
		return -ESRCH;
	task->make_it_fail = make_it_fail;
	put_task_struct(task);
	if (end - buffer == 0)
		return -EIO;
	return end - buffer;
}

865
static const struct file_operations proc_fault_inject_operations = {
866 867 868 869 870
	.read		= proc_fault_inject_read,
	.write		= proc_fault_inject_write,
};
#endif

I
Ingo Molnar 已提交
871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933
#ifdef CONFIG_SCHED_DEBUG
/*
 * Print out various scheduling related per-task fields:
 */
static int sched_show(struct seq_file *m, void *v)
{
	struct inode *inode = m->private;
	struct task_struct *p;

	WARN_ON(!inode);

	p = get_proc_task(inode);
	if (!p)
		return -ESRCH;
	proc_sched_show_task(p, m);

	put_task_struct(p);

	return 0;
}

static ssize_t
sched_write(struct file *file, const char __user *buf,
	    size_t count, loff_t *offset)
{
	struct inode *inode = file->f_path.dentry->d_inode;
	struct task_struct *p;

	WARN_ON(!inode);

	p = get_proc_task(inode);
	if (!p)
		return -ESRCH;
	proc_sched_set_task(p);

	put_task_struct(p);

	return count;
}

static int sched_open(struct inode *inode, struct file *filp)
{
	int ret;

	ret = single_open(filp, sched_show, NULL);
	if (!ret) {
		struct seq_file *m = filp->private_data;

		m->private = inode;
	}
	return ret;
}

static const struct file_operations proc_pid_sched_operations = {
	.open		= sched_open,
	.read		= seq_read,
	.write		= sched_write,
	.llseek		= seq_lseek,
	.release	= seq_release,
};

#endif

934
static void *proc_pid_follow_link(struct dentry *dentry, struct nameidata *nd)
L
Linus Torvalds 已提交
935 936 937 938 939 940 941
{
	struct inode *inode = dentry->d_inode;
	int error = -EACCES;

	/* We don't need a base pointer in the /proc filesystem */
	path_release(nd);

942 943
	/* Are we allowed to snoop on the tasks file descriptors? */
	if (!proc_fd_access_allowed(inode))
L
Linus Torvalds 已提交
944 945 946 947 948
		goto out;

	error = PROC_I(inode)->op.proc_get_link(inode, &nd->dentry, &nd->mnt);
	nd->last_type = LAST_BIND;
out:
949
	return ERR_PTR(error);
L
Linus Torvalds 已提交
950 951 952 953 954 955 956 957 958 959 960
}

static int do_proc_readlink(struct dentry *dentry, struct vfsmount *mnt,
			    char __user *buffer, int buflen)
{
	struct inode * inode;
	char *tmp = (char*)__get_free_page(GFP_KERNEL), *path;
	int len;

	if (!tmp)
		return -ENOMEM;
961

L
Linus Torvalds 已提交
962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984
	inode = dentry->d_inode;
	path = d_path(dentry, mnt, tmp, PAGE_SIZE);
	len = PTR_ERR(path);
	if (IS_ERR(path))
		goto out;
	len = tmp + PAGE_SIZE - 1 - path;

	if (len > buflen)
		len = buflen;
	if (copy_to_user(buffer, path, len))
		len = -EFAULT;
 out:
	free_page((unsigned long)tmp);
	return len;
}

static int proc_pid_readlink(struct dentry * dentry, char __user * buffer, int buflen)
{
	int error = -EACCES;
	struct inode *inode = dentry->d_inode;
	struct dentry *de;
	struct vfsmount *mnt = NULL;

985 986
	/* Are we allowed to snoop on the tasks file descriptors? */
	if (!proc_fd_access_allowed(inode))
L
Linus Torvalds 已提交
987 988 989 990 991 992 993 994 995 996 997 998 999
		goto out;

	error = PROC_I(inode)->op.proc_get_link(inode, &de, &mnt);
	if (error)
		goto out;

	error = do_proc_readlink(de, mnt, buffer, buflen);
	dput(de);
	mntput(mnt);
out:
	return error;
}

1000
static const struct inode_operations proc_pid_link_inode_operations = {
L
Linus Torvalds 已提交
1001
	.readlink	= proc_pid_readlink,
1002 1003
	.follow_link	= proc_pid_follow_link,
	.setattr	= proc_setattr,
L
Linus Torvalds 已提交
1004 1005
};

1006 1007 1008 1009

/* building an inode */

static int task_dumpable(struct task_struct *task)
L
Linus Torvalds 已提交
1010
{
1011 1012
	int dumpable = 0;
	struct mm_struct *mm;
L
Linus Torvalds 已提交
1013

1014 1015 1016 1017 1018 1019 1020 1021 1022
	task_lock(task);
	mm = task->mm;
	if (mm)
		dumpable = mm->dumpable;
	task_unlock(task);
	if(dumpable == 1)
		return 1;
	return 0;
}
L
Linus Torvalds 已提交
1023 1024


1025
static struct inode *proc_pid_make_inode(struct super_block * sb, struct task_struct *task)
1026 1027 1028
{
	struct inode * inode;
	struct proc_inode *ei;
L
Linus Torvalds 已提交
1029

1030
	/* We need a new inode */
L
Linus Torvalds 已提交
1031

1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043
	inode = new_inode(sb);
	if (!inode)
		goto out;

	/* Common stuff */
	ei = PROC_I(inode);
	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
	inode->i_op = &proc_def_inode_operations;

	/*
	 * grab the reference to task.
	 */
1044
	ei->pid = get_task_pid(task, PIDTYPE_PID);
1045 1046 1047 1048 1049 1050 1051 1052
	if (!ei->pid)
		goto out_unlock;

	inode->i_uid = 0;
	inode->i_gid = 0;
	if (task_dumpable(task)) {
		inode->i_uid = task->euid;
		inode->i_gid = task->egid;
L
Linus Torvalds 已提交
1053
	}
1054 1055
	security_task_to_inode(task, inode);

L
Linus Torvalds 已提交
1056
out:
1057 1058 1059 1060 1061
	return inode;

out_unlock:
	iput(inode);
	return NULL;
L
Linus Torvalds 已提交
1062 1063
}

1064
static int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
L
Linus Torvalds 已提交
1065 1066
{
	struct inode *inode = dentry->d_inode;
1067 1068
	struct task_struct *task;
	generic_fillattr(inode, stat);
L
Linus Torvalds 已提交
1069

1070 1071 1072 1073 1074 1075 1076 1077 1078
	rcu_read_lock();
	stat->uid = 0;
	stat->gid = 0;
	task = pid_task(proc_pid(inode), PIDTYPE_PID);
	if (task) {
		if ((inode->i_mode == (S_IFDIR|S_IRUGO|S_IXUGO)) ||
		    task_dumpable(task)) {
			stat->uid = task->euid;
			stat->gid = task->egid;
L
Linus Torvalds 已提交
1079 1080
		}
	}
1081
	rcu_read_unlock();
A
Alan Cox 已提交
1082
	return 0;
L
Linus Torvalds 已提交
1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093
}

/* dentry stuff */

/*
 *	Exceptional case: normally we are not allowed to unhash a busy
 * directory. In this case, however, we can do it - no aliasing problems
 * due to the way we treat inodes.
 *
 * Rewrite the inode's ownerships here because the owning task may have
 * performed a setuid(), etc.
1094 1095 1096 1097 1098 1099 1100
 *
 * Before the /proc/pid/status file was created the only way to read
 * the effective uid of a /process was to stat /proc/pid.  Reading
 * /proc/pid/status is slow enough that procps and other packages
 * kept stating /proc/pid.  To keep the rules in /proc simple I have
 * made this apply to all per process world readable and executable
 * directories.
L
Linus Torvalds 已提交
1101 1102 1103 1104
 */
static int pid_revalidate(struct dentry *dentry, struct nameidata *nd)
{
	struct inode *inode = dentry->d_inode;
1105 1106 1107 1108
	struct task_struct *task = get_proc_task(inode);
	if (task) {
		if ((inode->i_mode == (S_IFDIR|S_IRUGO|S_IXUGO)) ||
		    task_dumpable(task)) {
L
Linus Torvalds 已提交
1109 1110 1111 1112 1113 1114
			inode->i_uid = task->euid;
			inode->i_gid = task->egid;
		} else {
			inode->i_uid = 0;
			inode->i_gid = 0;
		}
L
Linus Torvalds 已提交
1115
		inode->i_mode &= ~(S_ISUID | S_ISGID);
L
Linus Torvalds 已提交
1116
		security_task_to_inode(task, inode);
1117
		put_task_struct(task);
L
Linus Torvalds 已提交
1118 1119 1120 1121 1122 1123
		return 1;
	}
	d_drop(dentry);
	return 0;
}

1124
static int pid_delete_dentry(struct dentry * dentry)
1125
{
1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140
	/* Is the task we represent dead?
	 * If so, then don't put the dentry on the lru list,
	 * kill it immediately.
	 */
	return !proc_pid(dentry->d_inode)->tasks[PIDTYPE_PID].first;
}

static struct dentry_operations pid_dentry_operations =
{
	.d_revalidate	= pid_revalidate,
	.d_delete	= pid_delete_dentry,
};

/* Lookups */

1141 1142
typedef struct dentry *instantiate_t(struct inode *, struct dentry *,
				struct task_struct *, const void *);
1143

1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155
/*
 * Fill a directory entry.
 *
 * If possible create the dcache entry and derive our inode number and
 * file type from dcache entry.
 *
 * Since all of the proc inode numbers are dynamically generated, the inode
 * numbers do not exist until the inode is cache.  This means creating the
 * the dcache entry in readdir is necessary to keep the inode numbers
 * reported by readdir in sync with the inode numbers reported
 * by stat.
 */
1156 1157
static int proc_fill_cache(struct file *filp, void *dirent, filldir_t filldir,
	char *name, int len,
1158
	instantiate_t instantiate, struct task_struct *task, const void *ptr)
1159
{
1160
	struct dentry *child, *dir = filp->f_path.dentry;
1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197
	struct inode *inode;
	struct qstr qname;
	ino_t ino = 0;
	unsigned type = DT_UNKNOWN;

	qname.name = name;
	qname.len  = len;
	qname.hash = full_name_hash(name, len);

	child = d_lookup(dir, &qname);
	if (!child) {
		struct dentry *new;
		new = d_alloc(dir, &qname);
		if (new) {
			child = instantiate(dir->d_inode, new, task, ptr);
			if (child)
				dput(new);
			else
				child = new;
		}
	}
	if (!child || IS_ERR(child) || !child->d_inode)
		goto end_instantiate;
	inode = child->d_inode;
	if (inode) {
		ino = inode->i_ino;
		type = inode->i_mode >> 12;
	}
	dput(child);
end_instantiate:
	if (!ino)
		ino = find_inode_number(dir, &qname);
	if (!ino)
		ino = 1;
	return filldir(dirent, name, len, filp->f_pos, ino, type);
}

1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219
static unsigned name_to_int(struct dentry *dentry)
{
	const char *name = dentry->d_name.name;
	int len = dentry->d_name.len;
	unsigned n = 0;

	if (len > 1 && *name == '0')
		goto out;
	while (len-- > 0) {
		unsigned c = *name++ - '0';
		if (c > 9)
			goto out;
		if (n >= (~0U-9)/10)
			goto out;
		n *= 10;
		n += c;
	}
	return n;
out:
	return ~0U;
}

M
Miklos Szeredi 已提交
1220 1221 1222 1223
#define PROC_FDINFO_MAX 64

static int proc_fd_info(struct inode *inode, struct dentry **dentry,
			struct vfsmount **mnt, char *info)
1224 1225 1226 1227 1228
{
	struct task_struct *task = get_proc_task(inode);
	struct files_struct *files = NULL;
	struct file *file;
	int fd = proc_fd(inode);
1229 1230

	if (task) {
1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241
		files = get_files_struct(task);
		put_task_struct(task);
	}
	if (files) {
		/*
		 * We are not taking a ref to the file structure, so we must
		 * hold ->file_lock.
		 */
		spin_lock(&files->file_lock);
		file = fcheck_files(files, fd);
		if (file) {
M
Miklos Szeredi 已提交
1242 1243 1244 1245 1246 1247 1248 1249 1250 1251
			if (mnt)
				*mnt = mntget(file->f_path.mnt);
			if (dentry)
				*dentry = dget(file->f_path.dentry);
			if (info)
				snprintf(info, PROC_FDINFO_MAX,
					 "pos:\t%lli\n"
					 "flags:\t0%o\n",
					 (long long) file->f_pos,
					 file->f_flags);
1252 1253 1254
			spin_unlock(&files->file_lock);
			put_files_struct(files);
			return 0;
1255
		}
1256 1257
		spin_unlock(&files->file_lock);
		put_files_struct(files);
1258
	}
1259
	return -ENOENT;
1260 1261
}

M
Miklos Szeredi 已提交
1262 1263 1264 1265 1266 1267
static int proc_fd_link(struct inode *inode, struct dentry **dentry,
			struct vfsmount **mnt)
{
	return proc_fd_info(inode, dentry, mnt, NULL);
}

L
Linus Torvalds 已提交
1268 1269 1270
static int tid_fd_revalidate(struct dentry *dentry, struct nameidata *nd)
{
	struct inode *inode = dentry->d_inode;
1271
	struct task_struct *task = get_proc_task(inode);
1272
	int fd = proc_fd(inode);
L
Linus Torvalds 已提交
1273 1274
	struct files_struct *files;

1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288
	if (task) {
		files = get_files_struct(task);
		if (files) {
			rcu_read_lock();
			if (fcheck_files(files, fd)) {
				rcu_read_unlock();
				put_files_struct(files);
				if (task_dumpable(task)) {
					inode->i_uid = task->euid;
					inode->i_gid = task->egid;
				} else {
					inode->i_uid = 0;
					inode->i_gid = 0;
				}
L
Linus Torvalds 已提交
1289
				inode->i_mode &= ~(S_ISUID | S_ISGID);
1290 1291 1292 1293
				security_task_to_inode(task, inode);
				put_task_struct(task);
				return 1;
			}
1294
			rcu_read_unlock();
L
Linus Torvalds 已提交
1295 1296
			put_files_struct(files);
		}
1297
		put_task_struct(task);
L
Linus Torvalds 已提交
1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308
	}
	d_drop(dentry);
	return 0;
}

static struct dentry_operations tid_fd_dentry_operations =
{
	.d_revalidate	= tid_fd_revalidate,
	.d_delete	= pid_delete_dentry,
};

1309
static struct dentry *proc_fd_instantiate(struct inode *dir,
1310
	struct dentry *dentry, struct task_struct *task, const void *ptr)
L
Linus Torvalds 已提交
1311
{
1312
	unsigned fd = *(const unsigned *)ptr;
1313 1314 1315 1316 1317
	struct file *file;
	struct files_struct *files;
 	struct inode *inode;
 	struct proc_inode *ei;
	struct dentry *error = ERR_PTR(-ENOENT);
L
Linus Torvalds 已提交
1318

1319
	inode = proc_pid_make_inode(dir->i_sb, task);
L
Linus Torvalds 已提交
1320 1321 1322
	if (!inode)
		goto out;
	ei = PROC_I(inode);
1323
	ei->fd = fd;
L
Linus Torvalds 已提交
1324 1325
	files = get_files_struct(task);
	if (!files)
1326
		goto out_iput;
L
Linus Torvalds 已提交
1327
	inode->i_mode = S_IFLNK;
1328 1329 1330 1331 1332 1333

	/*
	 * We are not taking a ref to the file structure, so we must
	 * hold ->file_lock.
	 */
	spin_lock(&files->file_lock);
L
Linus Torvalds 已提交
1334 1335
	file = fcheck_files(files, fd);
	if (!file)
1336
		goto out_unlock;
L
Linus Torvalds 已提交
1337 1338 1339 1340
	if (file->f_mode & 1)
		inode->i_mode |= S_IRUSR | S_IXUSR;
	if (file->f_mode & 2)
		inode->i_mode |= S_IWUSR | S_IXUSR;
1341
	spin_unlock(&files->file_lock);
L
Linus Torvalds 已提交
1342
	put_files_struct(files);
1343

L
Linus Torvalds 已提交
1344 1345 1346 1347 1348
	inode->i_op = &proc_pid_link_inode_operations;
	inode->i_size = 64;
	ei->op.proc_get_link = proc_fd_link;
	dentry->d_op = &tid_fd_dentry_operations;
	d_add(dentry, inode);
1349 1350
	/* Close the race of the process dying before we return the dentry */
	if (tid_fd_revalidate(dentry, NULL))
1351
		error = NULL;
L
Linus Torvalds 已提交
1352

1353 1354 1355
 out:
	return error;
out_unlock:
1356
	spin_unlock(&files->file_lock);
L
Linus Torvalds 已提交
1357
	put_files_struct(files);
1358
out_iput:
L
Linus Torvalds 已提交
1359
	iput(inode);
1360
	goto out;
L
Linus Torvalds 已提交
1361 1362
}

M
Miklos Szeredi 已提交
1363 1364 1365
static struct dentry *proc_lookupfd_common(struct inode *dir,
					   struct dentry *dentry,
					   instantiate_t instantiate)
1366 1367 1368 1369 1370 1371 1372 1373 1374 1375
{
	struct task_struct *task = get_proc_task(dir);
	unsigned fd = name_to_int(dentry);
	struct dentry *result = ERR_PTR(-ENOENT);

	if (!task)
		goto out_no_task;
	if (fd == ~0U)
		goto out;

M
Miklos Szeredi 已提交
1376
	result = instantiate(dir, dentry, task, &fd);
1377 1378 1379 1380 1381 1382
out:
	put_task_struct(task);
out_no_task:
	return result;
}

M
Miklos Szeredi 已提交
1383 1384
static int proc_readfd_common(struct file * filp, void * dirent,
			      filldir_t filldir, instantiate_t instantiate)
1385
{
1386
	struct dentry *dentry = filp->f_path.dentry;
1387 1388 1389 1390 1391 1392
	struct inode *inode = dentry->d_inode;
	struct task_struct *p = get_proc_task(inode);
	unsigned int fd, tid, ino;
	int retval;
	struct files_struct * files;
	struct fdtable *fdt;
L
Linus Torvalds 已提交
1393

1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419
	retval = -ENOENT;
	if (!p)
		goto out_no_task;
	retval = 0;
	tid = p->pid;

	fd = filp->f_pos;
	switch (fd) {
		case 0:
			if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR) < 0)
				goto out;
			filp->f_pos++;
		case 1:
			ino = parent_ino(dentry);
			if (filldir(dirent, "..", 2, 1, ino, DT_DIR) < 0)
				goto out;
			filp->f_pos++;
		default:
			files = get_files_struct(p);
			if (!files)
				goto out;
			rcu_read_lock();
			fdt = files_fdtable(files);
			for (fd = filp->f_pos-2;
			     fd < fdt->max_fds;
			     fd++, filp->f_pos++) {
M
Miklos Szeredi 已提交
1420 1421
				char name[PROC_NUMBUF];
				int len;
1422 1423 1424 1425 1426

				if (!fcheck_files(files, fd))
					continue;
				rcu_read_unlock();

M
Miklos Szeredi 已提交
1427 1428 1429 1430
				len = snprintf(name, sizeof(name), "%d", fd);
				if (proc_fill_cache(filp, dirent, filldir,
						    name, len, instantiate,
						    p, &fd) < 0) {
1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444
					rcu_read_lock();
					break;
				}
				rcu_read_lock();
			}
			rcu_read_unlock();
			put_files_struct(files);
	}
out:
	put_task_struct(p);
out_no_task:
	return retval;
}

M
Miklos Szeredi 已提交
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
static struct dentry *proc_lookupfd(struct inode *dir, struct dentry *dentry,
				    struct nameidata *nd)
{
	return proc_lookupfd_common(dir, dentry, proc_fd_instantiate);
}

static int proc_readfd(struct file *filp, void *dirent, filldir_t filldir)
{
	return proc_readfd_common(filp, dirent, filldir, proc_fd_instantiate);
}

static ssize_t proc_fdinfo_read(struct file *file, char __user *buf,
				      size_t len, loff_t *ppos)
{
	char tmp[PROC_FDINFO_MAX];
	int err = proc_fd_info(file->f_path.dentry->d_inode, NULL, NULL, tmp);
	if (!err)
		err = simple_read_from_buffer(buf, len, ppos, tmp, strlen(tmp));
	return err;
}

static const struct file_operations proc_fdinfo_file_operations = {
	.open		= nonseekable_open,
	.read		= proc_fdinfo_read,
};

1471
static const struct file_operations proc_fd_operations = {
1472 1473
	.read		= generic_read_dir,
	.readdir	= proc_readfd,
L
Linus Torvalds 已提交
1474 1475
};

1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492
/*
 * /proc/pid/fd needs a special permission handler so that a process can still
 * access /proc/self/fd after it has executed a setuid().
 */
static int proc_fd_permission(struct inode *inode, int mask,
				struct nameidata *nd)
{
	int rv;

	rv = generic_permission(inode, mask, NULL);
	if (rv == 0)
		return 0;
	if (task_pid(current) == proc_pid(inode))
		rv = 0;
	return rv;
}

L
Linus Torvalds 已提交
1493 1494 1495
/*
 * proc directories can do almost nothing..
 */
1496
static const struct inode_operations proc_fd_inode_operations = {
L
Linus Torvalds 已提交
1497
	.lookup		= proc_lookupfd,
1498
	.permission	= proc_fd_permission,
1499
	.setattr	= proc_setattr,
L
Linus Torvalds 已提交
1500 1501
};

M
Miklos Szeredi 已提交
1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553
static struct dentry *proc_fdinfo_instantiate(struct inode *dir,
	struct dentry *dentry, struct task_struct *task, const void *ptr)
{
	unsigned fd = *(unsigned *)ptr;
 	struct inode *inode;
 	struct proc_inode *ei;
	struct dentry *error = ERR_PTR(-ENOENT);

	inode = proc_pid_make_inode(dir->i_sb, task);
	if (!inode)
		goto out;
	ei = PROC_I(inode);
	ei->fd = fd;
	inode->i_mode = S_IFREG | S_IRUSR;
	inode->i_fop = &proc_fdinfo_file_operations;
	dentry->d_op = &tid_fd_dentry_operations;
	d_add(dentry, inode);
	/* Close the race of the process dying before we return the dentry */
	if (tid_fd_revalidate(dentry, NULL))
		error = NULL;

 out:
	return error;
}

static struct dentry *proc_lookupfdinfo(struct inode *dir,
					struct dentry *dentry,
					struct nameidata *nd)
{
	return proc_lookupfd_common(dir, dentry, proc_fdinfo_instantiate);
}

static int proc_readfdinfo(struct file *filp, void *dirent, filldir_t filldir)
{
	return proc_readfd_common(filp, dirent, filldir,
				  proc_fdinfo_instantiate);
}

static const struct file_operations proc_fdinfo_operations = {
	.read		= generic_read_dir,
	.readdir	= proc_readfdinfo,
};

/*
 * proc directories can do almost nothing..
 */
static const struct inode_operations proc_fdinfo_inode_operations = {
	.lookup		= proc_lookupfdinfo,
	.setattr	= proc_setattr,
};


1554
static struct dentry *proc_pident_instantiate(struct inode *dir,
1555
	struct dentry *dentry, struct task_struct *task, const void *ptr)
1556
{
1557
	const struct pid_entry *p = ptr;
1558 1559 1560 1561
	struct inode *inode;
	struct proc_inode *ei;
	struct dentry *error = ERR_PTR(-EINVAL);

1562
	inode = proc_pid_make_inode(dir->i_sb, task);
1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583
	if (!inode)
		goto out;

	ei = PROC_I(inode);
	inode->i_mode = p->mode;
	if (S_ISDIR(inode->i_mode))
		inode->i_nlink = 2;	/* Use getattr to fix if necessary */
	if (p->iop)
		inode->i_op = p->iop;
	if (p->fop)
		inode->i_fop = p->fop;
	ei->op = p->op;
	dentry->d_op = &pid_dentry_operations;
	d_add(dentry, inode);
	/* Close the race of the process dying before we return the dentry */
	if (pid_revalidate(dentry, NULL))
		error = NULL;
out:
	return error;
}

L
Linus Torvalds 已提交
1584 1585
static struct dentry *proc_pident_lookup(struct inode *dir, 
					 struct dentry *dentry,
1586
					 const struct pid_entry *ents,
1587
					 unsigned int nents)
L
Linus Torvalds 已提交
1588 1589
{
	struct inode *inode;
1590
	struct dentry *error;
1591
	struct task_struct *task = get_proc_task(dir);
1592
	const struct pid_entry *p, *last;
L
Linus Torvalds 已提交
1593

1594
	error = ERR_PTR(-ENOENT);
L
Linus Torvalds 已提交
1595 1596
	inode = NULL;

1597 1598
	if (!task)
		goto out_no_task;
L
Linus Torvalds 已提交
1599

1600 1601 1602 1603
	/*
	 * Yes, it does not scale. And it should not. Don't add
	 * new entries into /proc/<tgid>/ without very good reasons.
	 */
1604 1605
	last = &ents[nents - 1];
	for (p = ents; p <= last; p++) {
L
Linus Torvalds 已提交
1606 1607 1608 1609 1610
		if (p->len != dentry->d_name.len)
			continue;
		if (!memcmp(dentry->d_name.name, p->name, p->len))
			break;
	}
1611
	if (p > last)
L
Linus Torvalds 已提交
1612 1613
		goto out;

1614
	error = proc_pident_instantiate(dir, dentry, task, p);
L
Linus Torvalds 已提交
1615
out:
1616 1617
	put_task_struct(task);
out_no_task:
1618
	return error;
L
Linus Torvalds 已提交
1619 1620
}

1621 1622
static int proc_pident_fill_cache(struct file *filp, void *dirent,
	filldir_t filldir, struct task_struct *task, const struct pid_entry *p)
1623 1624 1625 1626 1627
{
	return proc_fill_cache(filp, dirent, filldir, p->name, p->len,
				proc_pident_instantiate, task, p);
}

1628 1629
static int proc_pident_readdir(struct file *filp,
		void *dirent, filldir_t filldir,
1630
		const struct pid_entry *ents, unsigned int nents)
1631 1632 1633
{
	int i;
	int pid;
1634
	struct dentry *dentry = filp->f_path.dentry;
1635 1636
	struct inode *inode = dentry->d_inode;
	struct task_struct *task = get_proc_task(inode);
1637
	const struct pid_entry *p, *last;
1638 1639 1640 1641 1642
	ino_t ino;
	int ret;

	ret = -ENOENT;
	if (!task)
1643
		goto out_no_task;
1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669

	ret = 0;
	pid = task->pid;
	i = filp->f_pos;
	switch (i) {
	case 0:
		ino = inode->i_ino;
		if (filldir(dirent, ".", 1, i, ino, DT_DIR) < 0)
			goto out;
		i++;
		filp->f_pos++;
		/* fall through */
	case 1:
		ino = parent_ino(dentry);
		if (filldir(dirent, "..", 2, i, ino, DT_DIR) < 0)
			goto out;
		i++;
		filp->f_pos++;
		/* fall through */
	default:
		i -= 2;
		if (i >= nents) {
			ret = 1;
			goto out;
		}
		p = ents + i;
1670 1671
		last = &ents[nents - 1];
		while (p <= last) {
1672
			if (proc_pident_fill_cache(filp, dirent, filldir, task, p) < 0)
1673 1674 1675 1676 1677 1678 1679 1680
				goto out;
			filp->f_pos++;
			p++;
		}
	}

	ret = 1;
out:
1681 1682
	put_task_struct(task);
out_no_task:
1683
	return ret;
L
Linus Torvalds 已提交
1684 1685
}

1686 1687 1688 1689
#ifdef CONFIG_SECURITY
static ssize_t proc_pid_attr_read(struct file * file, char __user * buf,
				  size_t count, loff_t *ppos)
{
1690
	struct inode * inode = file->f_path.dentry->d_inode;
1691
	char *p = NULL;
1692 1693 1694 1695
	ssize_t length;
	struct task_struct *task = get_proc_task(inode);

	if (!task)
1696
		return -ESRCH;
1697 1698

	length = security_getprocattr(task,
1699
				      (char*)file->f_path.dentry->d_name.name,
1700
				      &p);
1701
	put_task_struct(task);
1702 1703 1704
	if (length > 0)
		length = simple_read_from_buffer(buf, count, ppos, p, length);
	kfree(p);
1705
	return length;
L
Linus Torvalds 已提交
1706 1707
}

1708 1709 1710
static ssize_t proc_pid_attr_write(struct file * file, const char __user * buf,
				   size_t count, loff_t *ppos)
{
1711
	struct inode * inode = file->f_path.dentry->d_inode;
1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736
	char *page;
	ssize_t length;
	struct task_struct *task = get_proc_task(inode);

	length = -ESRCH;
	if (!task)
		goto out_no_task;
	if (count > PAGE_SIZE)
		count = PAGE_SIZE;

	/* No partial writes. */
	length = -EINVAL;
	if (*ppos != 0)
		goto out;

	length = -ENOMEM;
	page = (char*)__get_free_page(GFP_USER);
	if (!page)
		goto out;

	length = -EFAULT;
	if (copy_from_user(page, buf, count))
		goto out_free;

	length = security_setprocattr(task,
1737
				      (char*)file->f_path.dentry->d_name.name,
1738 1739 1740 1741 1742 1743 1744 1745 1746
				      (void*)page, count);
out_free:
	free_page((unsigned long) page);
out:
	put_task_struct(task);
out_no_task:
	return length;
}

1747
static const struct file_operations proc_pid_attr_operations = {
1748 1749 1750 1751
	.read		= proc_pid_attr_read,
	.write		= proc_pid_attr_write,
};

1752
static const struct pid_entry attr_dir_stuff[] = {
1753 1754 1755 1756 1757 1758
	REG("current",    S_IRUGO|S_IWUGO, pid_attr),
	REG("prev",       S_IRUGO,	   pid_attr),
	REG("exec",       S_IRUGO|S_IWUGO, pid_attr),
	REG("fscreate",   S_IRUGO|S_IWUGO, pid_attr),
	REG("keycreate",  S_IRUGO|S_IWUGO, pid_attr),
	REG("sockcreate", S_IRUGO|S_IWUGO, pid_attr),
1759 1760
};

1761
static int proc_attr_dir_readdir(struct file * filp,
1762 1763 1764
			     void * dirent, filldir_t filldir)
{
	return proc_pident_readdir(filp,dirent,filldir,
1765
				   attr_dir_stuff,ARRAY_SIZE(attr_dir_stuff));
1766 1767
}

1768
static const struct file_operations proc_attr_dir_operations = {
L
Linus Torvalds 已提交
1769
	.read		= generic_read_dir,
1770
	.readdir	= proc_attr_dir_readdir,
L
Linus Torvalds 已提交
1771 1772
};

1773
static struct dentry *proc_attr_dir_lookup(struct inode *dir,
1774 1775
				struct dentry *dentry, struct nameidata *nd)
{
1776 1777
	return proc_pident_lookup(dir, dentry,
				  attr_dir_stuff, ARRAY_SIZE(attr_dir_stuff));
1778 1779
}

1780
static const struct inode_operations proc_attr_dir_inode_operations = {
1781
	.lookup		= proc_attr_dir_lookup,
1782
	.getattr	= pid_getattr,
1783
	.setattr	= proc_setattr,
L
Linus Torvalds 已提交
1784 1785
};

1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805
#endif

/*
 * /proc/self:
 */
static int proc_self_readlink(struct dentry *dentry, char __user *buffer,
			      int buflen)
{
	char tmp[PROC_NUMBUF];
	sprintf(tmp, "%d", current->tgid);
	return vfs_readlink(dentry,buffer,buflen,tmp);
}

static void *proc_self_follow_link(struct dentry *dentry, struct nameidata *nd)
{
	char tmp[PROC_NUMBUF];
	sprintf(tmp, "%d", current->tgid);
	return ERR_PTR(vfs_follow_link(nd,tmp));
}

1806
static const struct inode_operations proc_self_inode_operations = {
1807 1808 1809 1810
	.readlink	= proc_self_readlink,
	.follow_link	= proc_self_follow_link,
};

1811 1812 1813 1814 1815 1816 1817
/*
 * proc base
 *
 * These are the directory entries in the root directory of /proc
 * that properly belong to the /proc filesystem, as they describe
 * describe something that is process related.
 */
1818
static const struct pid_entry proc_base_stuff[] = {
1819
	NOD("self", S_IFLNK|S_IRWXUGO,
1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845
		&proc_self_inode_operations, NULL, {}),
};

/*
 *	Exceptional case: normally we are not allowed to unhash a busy
 * directory. In this case, however, we can do it - no aliasing problems
 * due to the way we treat inodes.
 */
static int proc_base_revalidate(struct dentry *dentry, struct nameidata *nd)
{
	struct inode *inode = dentry->d_inode;
	struct task_struct *task = get_proc_task(inode);
	if (task) {
		put_task_struct(task);
		return 1;
	}
	d_drop(dentry);
	return 0;
}

static struct dentry_operations proc_base_dentry_operations =
{
	.d_revalidate	= proc_base_revalidate,
	.d_delete	= pid_delete_dentry,
};

1846
static struct dentry *proc_base_instantiate(struct inode *dir,
1847
	struct dentry *dentry, struct task_struct *task, const void *ptr)
1848
{
1849
	const struct pid_entry *p = ptr;
1850 1851
	struct inode *inode;
	struct proc_inode *ei;
1852
	struct dentry *error = ERR_PTR(-EINVAL);
1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866

	/* Allocate the inode */
	error = ERR_PTR(-ENOMEM);
	inode = new_inode(dir->i_sb);
	if (!inode)
		goto out;

	/* Initialize the inode */
	ei = PROC_I(inode);
	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;

	/*
	 * grab the reference to the task.
	 */
1867
	ei->pid = get_task_pid(task, PIDTYPE_PID);
1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892
	if (!ei->pid)
		goto out_iput;

	inode->i_uid = 0;
	inode->i_gid = 0;
	inode->i_mode = p->mode;
	if (S_ISDIR(inode->i_mode))
		inode->i_nlink = 2;
	if (S_ISLNK(inode->i_mode))
		inode->i_size = 64;
	if (p->iop)
		inode->i_op = p->iop;
	if (p->fop)
		inode->i_fop = p->fop;
	ei->op = p->op;
	dentry->d_op = &proc_base_dentry_operations;
	d_add(dentry, inode);
	error = NULL;
out:
	return error;
out_iput:
	iput(inode);
	goto out;
}

1893 1894 1895 1896
static struct dentry *proc_base_lookup(struct inode *dir, struct dentry *dentry)
{
	struct dentry *error;
	struct task_struct *task = get_proc_task(dir);
1897
	const struct pid_entry *p, *last;
1898 1899 1900 1901 1902 1903 1904

	error = ERR_PTR(-ENOENT);

	if (!task)
		goto out_no_task;

	/* Lookup the directory entry */
1905 1906
	last = &proc_base_stuff[ARRAY_SIZE(proc_base_stuff) - 1];
	for (p = proc_base_stuff; p <= last; p++) {
1907 1908 1909 1910 1911
		if (p->len != dentry->d_name.len)
			continue;
		if (!memcmp(dentry->d_name.name, p->name, p->len))
			break;
	}
1912
	if (p > last)
1913 1914 1915 1916 1917 1918 1919 1920 1921 1922
		goto out;

	error = proc_base_instantiate(dir, dentry, task, p);

out:
	put_task_struct(task);
out_no_task:
	return error;
}

1923 1924
static int proc_base_fill_cache(struct file *filp, void *dirent,
	filldir_t filldir, struct task_struct *task, const struct pid_entry *p)
1925 1926 1927 1928 1929
{
	return proc_fill_cache(filp, dirent, filldir, p->name, p->len,
				proc_base_instantiate, task, p);
}

1930 1931 1932 1933
#ifdef CONFIG_TASK_IO_ACCOUNTING
static int proc_pid_io_accounting(struct task_struct *task, char *buffer)
{
	return sprintf(buffer,
1934
#ifdef CONFIG_TASK_XACCT
1935 1936 1937 1938
			"rchar: %llu\n"
			"wchar: %llu\n"
			"syscr: %llu\n"
			"syscw: %llu\n"
1939
#endif
1940 1941 1942
			"read_bytes: %llu\n"
			"write_bytes: %llu\n"
			"cancelled_write_bytes: %llu\n",
1943
#ifdef CONFIG_TASK_XACCT
1944 1945 1946 1947
			(unsigned long long)task->rchar,
			(unsigned long long)task->wchar,
			(unsigned long long)task->syscr,
			(unsigned long long)task->syscw,
1948
#endif
1949 1950 1951 1952 1953 1954
			(unsigned long long)task->ioac.read_bytes,
			(unsigned long long)task->ioac.write_bytes,
			(unsigned long long)task->ioac.cancelled_write_bytes);
}
#endif

1955 1956 1957
/*
 * Thread groups
 */
1958
static const struct file_operations proc_task_operations;
1959
static const struct inode_operations proc_task_inode_operations;
1960

1961
static const struct pid_entry tgid_base_stuff[] = {
1962 1963
	DIR("task",       S_IRUGO|S_IXUGO, task),
	DIR("fd",         S_IRUSR|S_IXUSR, fd),
M
Miklos Szeredi 已提交
1964
	DIR("fdinfo",     S_IRUSR|S_IXUSR, fdinfo),
1965 1966 1967
	INF("environ",    S_IRUSR, pid_environ),
	INF("auxv",       S_IRUSR, pid_auxv),
	INF("status",     S_IRUGO, pid_status),
I
Ingo Molnar 已提交
1968 1969 1970
#ifdef CONFIG_SCHED_DEBUG
	REG("sched",      S_IRUGO|S_IWUSR, pid_sched),
#endif
1971 1972 1973 1974
	INF("cmdline",    S_IRUGO, pid_cmdline),
	INF("stat",       S_IRUGO, tgid_stat),
	INF("statm",      S_IRUGO, pid_statm),
	REG("maps",       S_IRUGO, maps),
1975
#ifdef CONFIG_NUMA
1976
	REG("numa_maps",  S_IRUGO, numa_maps),
1977
#endif
1978 1979 1980 1981 1982 1983
	REG("mem",        S_IRUSR|S_IWUSR, mem),
	LNK("cwd",        cwd),
	LNK("root",       root),
	LNK("exe",        exe),
	REG("mounts",     S_IRUGO, mounts),
	REG("mountstats", S_IRUSR, mountstats),
1984
#ifdef CONFIG_MMU
1985
	REG("clear_refs", S_IWUSR, clear_refs),
1986
	REG("smaps",      S_IRUGO, smaps),
1987 1988
#endif
#ifdef CONFIG_SECURITY
1989
	DIR("attr",       S_IRUGO|S_IXUGO, attr_dir),
1990 1991
#endif
#ifdef CONFIG_KALLSYMS
1992
	INF("wchan",      S_IRUGO, pid_wchan),
1993 1994
#endif
#ifdef CONFIG_SCHEDSTATS
1995
	INF("schedstat",  S_IRUGO, pid_schedstat),
1996 1997
#endif
#ifdef CONFIG_CPUSETS
1998
	REG("cpuset",     S_IRUGO, cpuset),
1999
#endif
2000 2001
	INF("oom_score",  S_IRUGO, oom_score),
	REG("oom_adj",    S_IRUGO|S_IWUSR, oom_adjust),
2002
#ifdef CONFIG_AUDITSYSCALL
2003
	REG("loginuid",   S_IWUSR|S_IRUGO, loginuid),
2004
#endif
2005 2006 2007
#ifdef CONFIG_FAULT_INJECTION
	REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject),
#endif
2008 2009 2010
#ifdef CONFIG_TASK_IO_ACCOUNTING
	INF("io",	S_IRUGO, pid_io_accounting),
#endif
2011
};
L
Linus Torvalds 已提交
2012

2013
static int proc_tgid_base_readdir(struct file * filp,
L
Linus Torvalds 已提交
2014 2015 2016
			     void * dirent, filldir_t filldir)
{
	return proc_pident_readdir(filp,dirent,filldir,
2017
				   tgid_base_stuff,ARRAY_SIZE(tgid_base_stuff));
L
Linus Torvalds 已提交
2018 2019
}

2020
static const struct file_operations proc_tgid_base_operations = {
L
Linus Torvalds 已提交
2021
	.read		= generic_read_dir,
2022
	.readdir	= proc_tgid_base_readdir,
L
Linus Torvalds 已提交
2023 2024
};

2025
static struct dentry *proc_tgid_base_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd){
2026 2027
	return proc_pident_lookup(dir, dentry,
				  tgid_base_stuff, ARRAY_SIZE(tgid_base_stuff));
L
Linus Torvalds 已提交
2028 2029
}

2030
static const struct inode_operations proc_tgid_base_inode_operations = {
2031
	.lookup		= proc_tgid_base_lookup,
2032
	.getattr	= pid_getattr,
2033
	.setattr	= proc_setattr,
L
Linus Torvalds 已提交
2034 2035 2036
};

/**
2037 2038 2039 2040 2041 2042 2043 2044 2045
 * proc_flush_task -  Remove dcache entries for @task from the /proc dcache.
 *
 * @task: task that should be flushed.
 *
 * Looks in the dcache for
 * /proc/@pid
 * /proc/@tgid/task/@pid
 * if either directory is present flushes it and all of it'ts children
 * from the dcache.
L
Linus Torvalds 已提交
2046
 *
2047 2048 2049 2050 2051
 * It is safe and reasonable to cache /proc entries for a task until
 * that task exits.  After that they just clog up the dcache with
 * useless entries, possibly causing useful dcache entries to be
 * flushed instead.  This routine is proved to flush those useless
 * dcache entries at process exit time.
L
Linus Torvalds 已提交
2052
 *
2053 2054 2055
 * NOTE: This routine is just an optimization so it does not guarantee
 *       that no dcache entries will exist at process exit time it
 *       just makes it very unlikely that any will persist.
L
Linus Torvalds 已提交
2056
 */
2057
void proc_flush_task(struct task_struct *task)
L
Linus Torvalds 已提交
2058
{
2059
	struct dentry *dentry, *leader, *dir;
2060
	char buf[PROC_NUMBUF];
2061 2062 2063 2064 2065 2066 2067 2068 2069 2070
	struct qstr name;

	name.name = buf;
	name.len = snprintf(buf, sizeof(buf), "%d", task->pid);
	dentry = d_hash_and_lookup(proc_mnt->mnt_root, &name);
	if (dentry) {
		shrink_dcache_parent(dentry);
		d_drop(dentry);
		dput(dentry);
	}
L
Linus Torvalds 已提交
2071

2072 2073
	if (thread_group_leader(task))
		goto out;
L
Linus Torvalds 已提交
2074

2075 2076 2077 2078 2079
	name.name = buf;
	name.len = snprintf(buf, sizeof(buf), "%d", task->tgid);
	leader = d_hash_and_lookup(proc_mnt->mnt_root, &name);
	if (!leader)
		goto out;
L
Linus Torvalds 已提交
2080

2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093
	name.name = "task";
	name.len = strlen(name.name);
	dir = d_hash_and_lookup(leader, &name);
	if (!dir)
		goto out_put_leader;

	name.name = buf;
	name.len = snprintf(buf, sizeof(buf), "%d", task->pid);
	dentry = d_hash_and_lookup(dir, &name);
	if (dentry) {
		shrink_dcache_parent(dentry);
		d_drop(dentry);
		dput(dentry);
L
Linus Torvalds 已提交
2094
	}
2095 2096 2097 2098 2099 2100

	dput(dir);
out_put_leader:
	dput(leader);
out:
	return;
L
Linus Torvalds 已提交
2101 2102
}

2103 2104
static struct dentry *proc_pid_instantiate(struct inode *dir,
					   struct dentry * dentry,
2105
					   struct task_struct *task, const void *ptr)
2106 2107 2108 2109
{
	struct dentry *error = ERR_PTR(-ENOENT);
	struct inode *inode;

2110
	inode = proc_pid_make_inode(dir->i_sb, task);
2111 2112 2113 2114 2115 2116 2117
	if (!inode)
		goto out;

	inode->i_mode = S_IFDIR|S_IRUGO|S_IXUGO;
	inode->i_op = &proc_tgid_base_inode_operations;
	inode->i_fop = &proc_tgid_base_operations;
	inode->i_flags|=S_IMMUTABLE;
M
Miklos Szeredi 已提交
2118
	inode->i_nlink = 5;
2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132
#ifdef CONFIG_SECURITY
	inode->i_nlink += 1;
#endif

	dentry->d_op = &pid_dentry_operations;

	d_add(dentry, inode);
	/* Close the race of the process dying before we return the dentry */
	if (pid_revalidate(dentry, NULL))
		error = NULL;
out:
	return error;
}

L
Linus Torvalds 已提交
2133 2134
struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *nd)
{
2135
	struct dentry *result = ERR_PTR(-ENOENT);
L
Linus Torvalds 已提交
2136 2137 2138
	struct task_struct *task;
	unsigned tgid;

2139 2140 2141 2142
	result = proc_base_lookup(dir, dentry);
	if (!IS_ERR(result) || PTR_ERR(result) != -ENOENT)
		goto out;

L
Linus Torvalds 已提交
2143 2144 2145 2146
	tgid = name_to_int(dentry);
	if (tgid == ~0U)
		goto out;

2147
	rcu_read_lock();
L
Linus Torvalds 已提交
2148 2149 2150
	task = find_task_by_pid(tgid);
	if (task)
		get_task_struct(task);
2151
	rcu_read_unlock();
L
Linus Torvalds 已提交
2152 2153 2154
	if (!task)
		goto out;

2155
	result = proc_pid_instantiate(dir, dentry, task, NULL);
L
Linus Torvalds 已提交
2156 2157
	put_task_struct(task);
out:
2158
	return result;
L
Linus Torvalds 已提交
2159 2160 2161
}

/*
2162
 * Find the first task with tgid >= tgid
2163
 *
L
Linus Torvalds 已提交
2164
 */
2165
static struct task_struct *next_tgid(unsigned int tgid)
L
Linus Torvalds 已提交
2166
{
2167 2168
	struct task_struct *task;
	struct pid *pid;
L
Linus Torvalds 已提交
2169

2170
	rcu_read_lock();
2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191
retry:
	task = NULL;
	pid = find_ge_pid(tgid);
	if (pid) {
		tgid = pid->nr + 1;
		task = pid_task(pid, PIDTYPE_PID);
		/* What we to know is if the pid we have find is the
		 * pid of a thread_group_leader.  Testing for task
		 * being a thread_group_leader is the obvious thing
		 * todo but there is a window when it fails, due to
		 * the pid transfer logic in de_thread.
		 *
		 * So we perform the straight forward test of seeing
		 * if the pid we have found is the pid of a thread
		 * group leader, and don't worry if the task we have
		 * found doesn't happen to be a thread group leader.
		 * As we don't care in the case of readdir.
		 */
		if (!task || !has_group_leader_pid(task))
			goto retry;
		get_task_struct(task);
2192
	}
2193
	rcu_read_unlock();
2194
	return task;
L
Linus Torvalds 已提交
2195 2196
}

2197
#define TGID_OFFSET (FIRST_PROCESS_ENTRY + ARRAY_SIZE(proc_base_stuff))
2198

2199 2200 2201 2202 2203 2204 2205 2206 2207
static int proc_pid_fill_cache(struct file *filp, void *dirent, filldir_t filldir,
	struct task_struct *task, int tgid)
{
	char name[PROC_NUMBUF];
	int len = snprintf(name, sizeof(name), "%d", tgid);
	return proc_fill_cache(filp, dirent, filldir, name, len,
				proc_pid_instantiate, task, NULL);
}

L
Linus Torvalds 已提交
2208 2209 2210 2211
/* for the /proc/ directory itself, after non-process stuff has been done */
int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir)
{
	unsigned int nr = filp->f_pos - FIRST_PROCESS_ENTRY;
2212
	struct task_struct *reaper = get_proc_task(filp->f_path.dentry->d_inode);
2213 2214
	struct task_struct *task;
	int tgid;
L
Linus Torvalds 已提交
2215

2216 2217 2218
	if (!reaper)
		goto out_no_task;

2219
	for (; nr < ARRAY_SIZE(proc_base_stuff); filp->f_pos++, nr++) {
2220
		const struct pid_entry *p = &proc_base_stuff[nr];
2221
		if (proc_base_fill_cache(filp, dirent, filldir, reaper, p) < 0)
2222
			goto out;
L
Linus Torvalds 已提交
2223 2224
	}

2225 2226
	tgid = filp->f_pos - TGID_OFFSET;
	for (task = next_tgid(tgid);
2227
	     task;
2228
	     put_task_struct(task), task = next_tgid(tgid + 1)) {
2229
		tgid = task->pid;
2230
		filp->f_pos = tgid + TGID_OFFSET;
2231
		if (proc_pid_fill_cache(filp, dirent, filldir, task, tgid) < 0) {
2232
			put_task_struct(task);
2233
			goto out;
L
Linus Torvalds 已提交
2234
		}
2235
	}
2236 2237
	filp->f_pos = PID_MAX_LIMIT + TGID_OFFSET;
out:
2238 2239
	put_task_struct(reaper);
out_no_task:
2240 2241
	return 0;
}
L
Linus Torvalds 已提交
2242

2243 2244 2245
/*
 * Tasks
 */
2246
static const struct pid_entry tid_base_stuff[] = {
2247
	DIR("fd",        S_IRUSR|S_IXUSR, fd),
M
Miklos Szeredi 已提交
2248
	DIR("fdinfo",    S_IRUSR|S_IXUSR, fdinfo),
2249 2250 2251
	INF("environ",   S_IRUSR, pid_environ),
	INF("auxv",      S_IRUSR, pid_auxv),
	INF("status",    S_IRUGO, pid_status),
I
Ingo Molnar 已提交
2252 2253 2254
#ifdef CONFIG_SCHED_DEBUG
	REG("sched",     S_IRUGO|S_IWUSR, pid_sched),
#endif
2255 2256 2257 2258
	INF("cmdline",   S_IRUGO, pid_cmdline),
	INF("stat",      S_IRUGO, tid_stat),
	INF("statm",     S_IRUGO, pid_statm),
	REG("maps",      S_IRUGO, maps),
2259
#ifdef CONFIG_NUMA
2260
	REG("numa_maps", S_IRUGO, numa_maps),
2261
#endif
2262 2263 2264 2265 2266
	REG("mem",       S_IRUSR|S_IWUSR, mem),
	LNK("cwd",       cwd),
	LNK("root",      root),
	LNK("exe",       exe),
	REG("mounts",    S_IRUGO, mounts),
2267
#ifdef CONFIG_MMU
2268
	REG("clear_refs", S_IWUSR, clear_refs),
2269
	REG("smaps",     S_IRUGO, smaps),
2270 2271
#endif
#ifdef CONFIG_SECURITY
2272
	DIR("attr",      S_IRUGO|S_IXUGO, attr_dir),
2273 2274
#endif
#ifdef CONFIG_KALLSYMS
2275
	INF("wchan",     S_IRUGO, pid_wchan),
2276 2277
#endif
#ifdef CONFIG_SCHEDSTATS
2278
	INF("schedstat", S_IRUGO, pid_schedstat),
2279 2280
#endif
#ifdef CONFIG_CPUSETS
2281
	REG("cpuset",    S_IRUGO, cpuset),
2282
#endif
2283 2284
	INF("oom_score", S_IRUGO, oom_score),
	REG("oom_adj",   S_IRUGO|S_IWUSR, oom_adjust),
2285
#ifdef CONFIG_AUDITSYSCALL
2286
	REG("loginuid",  S_IWUSR|S_IRUGO, loginuid),
2287
#endif
2288 2289 2290
#ifdef CONFIG_FAULT_INJECTION
	REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject),
#endif
2291 2292 2293 2294 2295 2296 2297 2298 2299 2300
};

static int proc_tid_base_readdir(struct file * filp,
			     void * dirent, filldir_t filldir)
{
	return proc_pident_readdir(filp,dirent,filldir,
				   tid_base_stuff,ARRAY_SIZE(tid_base_stuff));
}

static struct dentry *proc_tid_base_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd){
2301 2302
	return proc_pident_lookup(dir, dentry,
				  tid_base_stuff, ARRAY_SIZE(tid_base_stuff));
2303 2304
}

2305
static const struct file_operations proc_tid_base_operations = {
2306 2307 2308 2309
	.read		= generic_read_dir,
	.readdir	= proc_tid_base_readdir,
};

2310
static const struct inode_operations proc_tid_base_inode_operations = {
2311 2312 2313 2314 2315
	.lookup		= proc_tid_base_lookup,
	.getattr	= pid_getattr,
	.setattr	= proc_setattr,
};

2316
static struct dentry *proc_task_instantiate(struct inode *dir,
2317
	struct dentry *dentry, struct task_struct *task, const void *ptr)
2318 2319 2320
{
	struct dentry *error = ERR_PTR(-ENOENT);
	struct inode *inode;
2321
	inode = proc_pid_make_inode(dir->i_sb, task);
2322 2323 2324 2325 2326 2327 2328

	if (!inode)
		goto out;
	inode->i_mode = S_IFDIR|S_IRUGO|S_IXUGO;
	inode->i_op = &proc_tid_base_inode_operations;
	inode->i_fop = &proc_tid_base_operations;
	inode->i_flags|=S_IMMUTABLE;
M
Miklos Szeredi 已提交
2329
	inode->i_nlink = 4;
2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343
#ifdef CONFIG_SECURITY
	inode->i_nlink += 1;
#endif

	dentry->d_op = &pid_dentry_operations;

	d_add(dentry, inode);
	/* Close the race of the process dying before we return the dentry */
	if (pid_revalidate(dentry, NULL))
		error = NULL;
out:
	return error;
}

2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367
static struct dentry *proc_task_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *nd)
{
	struct dentry *result = ERR_PTR(-ENOENT);
	struct task_struct *task;
	struct task_struct *leader = get_proc_task(dir);
	unsigned tid;

	if (!leader)
		goto out_no_task;

	tid = name_to_int(dentry);
	if (tid == ~0U)
		goto out;

	rcu_read_lock();
	task = find_task_by_pid(tid);
	if (task)
		get_task_struct(task);
	rcu_read_unlock();
	if (!task)
		goto out;
	if (leader->tgid != task->tgid)
		goto out_drop_task;

2368
	result = proc_task_instantiate(dir, dentry, task, NULL);
2369 2370 2371 2372 2373 2374 2375 2376
out_drop_task:
	put_task_struct(task);
out:
	put_task_struct(leader);
out_no_task:
	return result;
}

2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388
/*
 * Find the first tid of a thread group to return to user space.
 *
 * Usually this is just the thread group leader, but if the users
 * buffer was too small or there was a seek into the middle of the
 * directory we have more work todo.
 *
 * In the case of a short read we start with find_task_by_pid.
 *
 * In the case of a seek we start with the leader and walk nr
 * threads past it.
 */
2389 2390
static struct task_struct *first_tid(struct task_struct *leader,
					int tid, int nr)
2391
{
O
Oleg Nesterov 已提交
2392
	struct task_struct *pos;
L
Linus Torvalds 已提交
2393

2394
	rcu_read_lock();
2395 2396 2397
	/* Attempt to start with the pid of a thread */
	if (tid && (nr > 0)) {
		pos = find_task_by_pid(tid);
O
Oleg Nesterov 已提交
2398 2399
		if (pos && (pos->group_leader == leader))
			goto found;
2400
	}
L
Linus Torvalds 已提交
2401

2402
	/* If nr exceeds the number of threads there is nothing todo */
O
Oleg Nesterov 已提交
2403 2404 2405
	pos = NULL;
	if (nr && nr >= get_nr_threads(leader))
		goto out;
L
Linus Torvalds 已提交
2406

O
Oleg Nesterov 已提交
2407 2408
	/* If we haven't found our starting place yet start
	 * with the leader and walk nr threads forward.
2409
	 */
O
Oleg Nesterov 已提交
2410 2411 2412 2413 2414 2415
	for (pos = leader; nr > 0; --nr) {
		pos = next_thread(pos);
		if (pos == leader) {
			pos = NULL;
			goto out;
		}
L
Linus Torvalds 已提交
2416
	}
O
Oleg Nesterov 已提交
2417 2418 2419
found:
	get_task_struct(pos);
out:
2420
	rcu_read_unlock();
2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431
	return pos;
}

/*
 * Find the next thread in the thread list.
 * Return NULL if there is an error or no next thread.
 *
 * The reference to the input task_struct is released.
 */
static struct task_struct *next_tid(struct task_struct *start)
{
O
Oleg Nesterov 已提交
2432
	struct task_struct *pos = NULL;
2433
	rcu_read_lock();
O
Oleg Nesterov 已提交
2434
	if (pid_alive(start)) {
2435
		pos = next_thread(start);
O
Oleg Nesterov 已提交
2436 2437 2438 2439 2440
		if (thread_group_leader(pos))
			pos = NULL;
		else
			get_task_struct(pos);
	}
2441
	rcu_read_unlock();
2442 2443
	put_task_struct(start);
	return pos;
L
Linus Torvalds 已提交
2444 2445
}

2446 2447 2448 2449 2450 2451 2452 2453 2454
static int proc_task_fill_cache(struct file *filp, void *dirent, filldir_t filldir,
	struct task_struct *task, int tid)
{
	char name[PROC_NUMBUF];
	int len = snprintf(name, sizeof(name), "%d", tid);
	return proc_fill_cache(filp, dirent, filldir, name, len,
				proc_task_instantiate, task, NULL);
}

L
Linus Torvalds 已提交
2455 2456 2457
/* for the /proc/TGID/task/ directories */
static int proc_task_readdir(struct file * filp, void * dirent, filldir_t filldir)
{
2458
	struct dentry *dentry = filp->f_path.dentry;
L
Linus Torvalds 已提交
2459
	struct inode *inode = dentry->d_inode;
2460
	struct task_struct *leader = NULL;
2461
	struct task_struct *task;
L
Linus Torvalds 已提交
2462 2463
	int retval = -ENOENT;
	ino_t ino;
2464
	int tid;
L
Linus Torvalds 已提交
2465 2466
	unsigned long pos = filp->f_pos;  /* avoiding "long long" filp->f_pos */

2467 2468 2469 2470 2471 2472 2473 2474 2475 2476
	task = get_proc_task(inode);
	if (!task)
		goto out_no_task;
	rcu_read_lock();
	if (pid_alive(task)) {
		leader = task->group_leader;
		get_task_struct(leader);
	}
	rcu_read_unlock();
	put_task_struct(task);
2477 2478
	if (!leader)
		goto out_no_task;
L
Linus Torvalds 已提交
2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495
	retval = 0;

	switch (pos) {
	case 0:
		ino = inode->i_ino;
		if (filldir(dirent, ".", 1, pos, ino, DT_DIR) < 0)
			goto out;
		pos++;
		/* fall through */
	case 1:
		ino = parent_ino(dentry);
		if (filldir(dirent, "..", 2, pos, ino, DT_DIR) < 0)
			goto out;
		pos++;
		/* fall through */
	}

2496 2497 2498 2499 2500 2501 2502 2503 2504
	/* f_version caches the tgid value that the last readdir call couldn't
	 * return. lseek aka telldir automagically resets f_version to 0.
	 */
	tid = filp->f_version;
	filp->f_version = 0;
	for (task = first_tid(leader, tid, pos - 2);
	     task;
	     task = next_tid(task), pos++) {
		tid = task->pid;
2505
		if (proc_task_fill_cache(filp, dirent, filldir, task, tid) < 0) {
2506 2507 2508 2509
			/* returning this tgid failed, save it as the first
			 * pid for the next readir call */
			filp->f_version = tid;
			put_task_struct(task);
L
Linus Torvalds 已提交
2510
			break;
2511
		}
L
Linus Torvalds 已提交
2512 2513 2514
	}
out:
	filp->f_pos = pos;
2515 2516
	put_task_struct(leader);
out_no_task:
L
Linus Torvalds 已提交
2517 2518
	return retval;
}
2519 2520 2521 2522

static int proc_task_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
{
	struct inode *inode = dentry->d_inode;
2523
	struct task_struct *p = get_proc_task(inode);
2524 2525
	generic_fillattr(inode, stat);

2526 2527 2528 2529 2530
	if (p) {
		rcu_read_lock();
		stat->nlink += get_nr_threads(p);
		rcu_read_unlock();
		put_task_struct(p);
2531 2532 2533 2534
	}

	return 0;
}
2535

2536
static const struct inode_operations proc_task_inode_operations = {
2537 2538 2539 2540 2541
	.lookup		= proc_task_lookup,
	.getattr	= proc_task_getattr,
	.setattr	= proc_setattr,
};

2542
static const struct file_operations proc_task_operations = {
2543 2544 2545
	.read		= generic_read_dir,
	.readdir	= proc_task_readdir,
};