base.c 60.1 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 72
#include <linux/mount.h>
#include <linux/security.h>
#include <linux/ptrace.h>
#include <linux/seccomp.h>
#include <linux/cpuset.h>
#include <linux/audit.h>
A
Al Viro 已提交
73
#include <linux/poll.h>
74
#include <linux/nsproxy.h>
75
#include <linux/oom.h>
L
Linus Torvalds 已提交
76 77
#include "internal.h"

78 79 80 81 82 83 84 85 86 87
/* 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
88

89
/* Worst case buffer size needed for holding an integer. */
A
Andrew Morton 已提交
90
#define PROC_NUMBUF 13
91

L
Linus Torvalds 已提交
92 93
struct pid_entry {
	char *name;
94
	int len;
L
Linus Torvalds 已提交
95
	mode_t mode;
96
	const struct inode_operations *iop;
97
	const struct file_operations *fop;
98
	union proc_op op;
L
Linus Torvalds 已提交
99 100
};

101
#define NOD(NAME, MODE, IOP, FOP, OP) {			\
102
	.name = (NAME),					\
103
	.len  = sizeof(NAME) - 1,			\
104 105 106 107 108 109
	.mode = MODE,					\
	.iop  = IOP,					\
	.fop  = FOP,					\
	.op   = OP,					\
}

110 111
#define DIR(NAME, MODE, OTYPE)							\
	NOD(NAME, (S_IFDIR|(MODE)),						\
112 113
		&proc_##OTYPE##_inode_operations, &proc_##OTYPE##_operations,	\
		{} )
114 115
#define LNK(NAME, OTYPE)					\
	NOD(NAME, (S_IFLNK|S_IRWXUGO),				\
116 117
		&proc_pid_link_inode_operations, NULL,		\
		{ .proc_get_link = &proc_##OTYPE##_link } )
118 119
#define REG(NAME, MODE, OTYPE)				\
	NOD(NAME, (S_IFREG|(MODE)), NULL,		\
120
		&proc_##OTYPE##_operations, {})
121 122
#define INF(NAME, MODE, OTYPE)				\
	NOD(NAME, (S_IFREG|(MODE)), 			\
123 124
		NULL, &proc_info_file_operations,	\
		{ .proc_read = &proc_##OTYPE } )
L
Linus Torvalds 已提交
125

K
Kees Cook 已提交
126 127 128
int maps_protect;
EXPORT_SYMBOL(maps_protect);

129
static struct fs_struct *get_fs_struct(struct task_struct *task)
L
Linus Torvalds 已提交
130 131
{
	struct fs_struct *fs;
132 133
	task_lock(task);
	fs = task->fs;
L
Linus Torvalds 已提交
134 135
	if(fs)
		atomic_inc(&fs->count);
136 137 138 139
	task_unlock(task);
	return fs;
}

140 141 142 143 144 145 146 147 148 149 150 151 152
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;
}

153 154
static int proc_cwd_link(struct inode *inode, struct dentry **dentry, struct vfsmount **mnt)
{
155 156
	struct task_struct *task = get_proc_task(inode);
	struct fs_struct *fs = NULL;
157
	int result = -ENOENT;
158 159 160 161 162

	if (task) {
		fs = get_fs_struct(task);
		put_task_struct(task);
	}
L
Linus Torvalds 已提交
163 164 165 166 167 168 169 170 171 172 173 174 175
	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)
{
176 177
	struct task_struct *task = get_proc_task(inode);
	struct fs_struct *fs = NULL;
L
Linus Torvalds 已提交
178
	int result = -ENOENT;
179 180 181 182 183

	if (task) {
		fs = get_fs_struct(task);
		put_task_struct(task);
	}
L
Linus Torvalds 已提交
184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210
	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) {
		unsigned int len = mm->env_end - mm->env_start;
		if (len > PAGE_SIZE)
			len = PAGE_SIZE;
		res = access_process_vm(task, mm->env_start, buffer, len, 0);
211
		if (!ptrace_may_attach(task))
L
Linus Torvalds 已提交
212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315
			res = -ESRCH;
		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)
{
	char *modname;
	const char *sym_name;
	unsigned long wchan, size, offset;
	char namebuf[KSYM_NAME_LEN+1];

	wchan = get_wchan(task);

	sym_name = kallsyms_lookup(wchan, &size, &offset, &modname, namebuf);
	if (sym_name)
		return sprintf(buffer, "%s", sym_name);
	return sprintf(buffer, "%lu", wchan);
}
#endif /* CONFIG_KALLSYMS */

#ifdef CONFIG_SCHEDSTATS
/*
 * Provides /proc/PID/schedstat
 */
static int proc_pid_schedstat(struct task_struct *task, char *buffer)
{
	return sprintf(buffer, "%lu %lu %lu\n",
			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);
316
	read_lock(&tasklist_lock);
L
Linus Torvalds 已提交
317
	points = badness(task, uptime.tv_sec);
318
	read_unlock(&tasklist_lock);
L
Linus Torvalds 已提交
319 320 321 322 323 324 325 326
	return sprintf(buffer, "%lu\n", points);
}

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

/* permission checks */
327
static int proc_fd_access_allowed(struct inode *inode)
L
Linus Torvalds 已提交
328
{
329 330
	struct task_struct *task;
	int allowed = 0;
331 332 333
	/* 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.
334 335
	 */
	task = get_proc_task(inode);
336 337
	if (task) {
		allowed = ptrace_may_attach(task);
338
		put_task_struct(task);
339
	}
340
	return allowed;
L
Linus Torvalds 已提交
341 342
}

343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359
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);
	if (!error) {
		error = security_inode_setattr(dentry, attr);
		if (!error)
			error = inode_setattr(inode, attr);
	}
	return error;
}

360
static const struct inode_operations proc_def_inode_operations = {
361 362 363
	.setattr	= proc_setattr,
};

L
Linus Torvalds 已提交
364
extern struct seq_operations mounts_op;
A
Al Viro 已提交
365 366 367 368 369
struct proc_mounts {
	struct seq_file m;
	int event;
};

L
Linus Torvalds 已提交
370 371
static int mounts_open(struct inode *inode, struct file *file)
{
372
	struct task_struct *task = get_proc_task(inode);
373
	struct mnt_namespace *ns = NULL;
A
Al Viro 已提交
374 375
	struct proc_mounts *p;
	int ret = -EINVAL;
L
Linus Torvalds 已提交
376

377 378
	if (task) {
		task_lock(task);
379 380 381 382 383
		if (task->nsproxy) {
			ns = task->nsproxy->mnt_ns;
			if (ns)
				get_mnt_ns(ns);
		}
384 385 386
		task_unlock(task);
		put_task_struct(task);
	}
A
Al Viro 已提交
387

388
	if (ns) {
A
Al Viro 已提交
389 390 391 392 393 394
		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) {
395 396
				p->m.private = ns;
				p->event = ns->event;
A
Al Viro 已提交
397 398 399
				return 0;
			}
			kfree(p);
L
Linus Torvalds 已提交
400
		}
401
		put_mnt_ns(ns);
L
Linus Torvalds 已提交
402 403 404 405 406 407 408
	}
	return ret;
}

static int mounts_release(struct inode *inode, struct file *file)
{
	struct seq_file *m = file->private_data;
409 410
	struct mnt_namespace *ns = m->private;
	put_mnt_ns(ns);
L
Linus Torvalds 已提交
411 412 413
	return seq_release(inode, file);
}

A
Al Viro 已提交
414 415 416
static unsigned mounts_poll(struct file *file, poll_table *wait)
{
	struct proc_mounts *p = file->private_data;
417
	struct mnt_namespace *ns = p->m.private;
A
Al Viro 已提交
418 419 420 421 422 423 424 425 426 427 428 429 430 431
	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;
}

432
static const struct file_operations proc_mounts_operations = {
L
Linus Torvalds 已提交
433 434 435 436
	.open		= mounts_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
	.release	= mounts_release,
A
Al Viro 已提交
437
	.poll		= mounts_poll,
L
Linus Torvalds 已提交
438 439
};

440 441 442 443 444 445 446
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;
447
		struct mnt_namespace *mnt_ns = NULL;
448 449 450 451
		struct task_struct *task = get_proc_task(inode);

		if (task) {
			task_lock(task);
452
			if (task->nsproxy)
453 454 455
				mnt_ns = task->nsproxy->mnt_ns;
			if (mnt_ns)
				get_mnt_ns(mnt_ns);
456 457 458
			task_unlock(task);
			put_task_struct(task);
		}
459

460 461
		if (mnt_ns)
			m->private = mnt_ns;
462 463 464 465 466 467 468 469
		else {
			seq_release(inode, file);
			ret = -EINVAL;
		}
	}
	return ret;
}

470
static const struct file_operations proc_mountstats_operations = {
471 472 473 474 475 476
	.open		= mountstats_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
	.release	= mounts_release,
};

L
Linus Torvalds 已提交
477 478 479 480 481
#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)
{
482
	struct inode * inode = file->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
483 484
	unsigned long page;
	ssize_t length;
485 486 487 488 489
	struct task_struct *task = get_proc_task(inode);

	length = -ESRCH;
	if (!task)
		goto out_no_task;
L
Linus Torvalds 已提交
490 491 492

	if (count > PROC_BLOCK_SIZE)
		count = PROC_BLOCK_SIZE;
493 494

	length = -ENOMEM;
L
Linus Torvalds 已提交
495
	if (!(page = __get_free_page(GFP_KERNEL)))
496
		goto out;
L
Linus Torvalds 已提交
497 498 499 500 501 502

	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);
503 504 505
out:
	put_task_struct(task);
out_no_task:
L
Linus Torvalds 已提交
506 507 508
	return length;
}

509
static const struct file_operations proc_info_file_operations = {
L
Linus Torvalds 已提交
510 511 512 513 514 515 516 517 518 519 520 521
	.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)
{
522
	struct task_struct *task = get_proc_task(file->f_path.dentry->d_inode);
L
Linus Torvalds 已提交
523 524 525 526 527
	char *page;
	unsigned long src = *ppos;
	int ret = -ESRCH;
	struct mm_struct *mm;

528 529 530
	if (!task)
		goto out_no_task;

531
	if (!MAY_PTRACE(task) || !ptrace_may_attach(task))
L
Linus Torvalds 已提交
532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556
		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);
557
		if (!retval || !MAY_PTRACE(task) || !ptrace_may_attach(task)) {
L
Linus Torvalds 已提交
558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579
			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:
580 581
	put_task_struct(task);
out_no_task:
L
Linus Torvalds 已提交
582 583 584 585 586 587 588
	return ret;
}

#define mem_write NULL

#ifndef mem_write
/* This is a security hazard */
589
static ssize_t mem_write(struct file * file, const char __user *buf,
L
Linus Torvalds 已提交
590 591
			 size_t count, loff_t *ppos)
{
592
	int copied;
L
Linus Torvalds 已提交
593
	char *page;
594
	struct task_struct *task = get_proc_task(file->f_path.dentry->d_inode);
L
Linus Torvalds 已提交
595 596
	unsigned long dst = *ppos;

597 598 599 600
	copied = -ESRCH;
	if (!task)
		goto out_no_task;

601
	if (!MAY_PTRACE(task) || !ptrace_may_attach(task))
602
		goto out;
L
Linus Torvalds 已提交
603

604
	copied = -ENOMEM;
L
Linus Torvalds 已提交
605 606
	page = (char *)__get_free_page(GFP_USER);
	if (!page)
607
		goto out;
L
Linus Torvalds 已提交
608

609
	copied = 0;
L
Linus Torvalds 已提交
610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630
	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);
631 632 633
out:
	put_task_struct(task);
out_no_task:
L
Linus Torvalds 已提交
634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653
	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;
}

654
static const struct file_operations proc_mem_operations = {
L
Linus Torvalds 已提交
655 656 657 658 659 660 661 662 663
	.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)
{
664
	struct task_struct *task = get_proc_task(file->f_path.dentry->d_inode);
665
	char buffer[PROC_NUMBUF];
L
Linus Torvalds 已提交
666
	size_t len;
667
	int oom_adjust;
L
Linus Torvalds 已提交
668 669
	loff_t __ppos = *ppos;

670 671 672 673 674
	if (!task)
		return -ESRCH;
	oom_adjust = task->oomkilladj;
	put_task_struct(task);

675
	len = snprintf(buffer, sizeof(buffer), "%i\n", oom_adjust);
L
Linus Torvalds 已提交
676 677 678 679 680 681 682 683 684 685 686 687 688
	if (__ppos >= len)
		return 0;
	if (count > len-__ppos)
		count = len-__ppos;
	if (copy_to_user(buf, buffer + __ppos, count))
		return -EFAULT;
	*ppos = __ppos + count;
	return count;
}

static ssize_t oom_adjust_write(struct file *file, const char __user *buf,
				size_t count, loff_t *ppos)
{
689
	struct task_struct *task;
690
	char buffer[PROC_NUMBUF], *end;
L
Linus Torvalds 已提交
691 692
	int oom_adjust;

693 694 695
	memset(buffer, 0, sizeof(buffer));
	if (count > sizeof(buffer) - 1)
		count = sizeof(buffer) - 1;
L
Linus Torvalds 已提交
696 697 698
	if (copy_from_user(buffer, buf, count))
		return -EFAULT;
	oom_adjust = simple_strtol(buffer, &end, 0);
699 700
	if ((oom_adjust < OOM_ADJUST_MIN || oom_adjust > OOM_ADJUST_MAX) &&
	     oom_adjust != OOM_DISABLE)
L
Linus Torvalds 已提交
701 702 703
		return -EINVAL;
	if (*end == '\n')
		end++;
704
	task = get_proc_task(file->f_path.dentry->d_inode);
705 706
	if (!task)
		return -ESRCH;
707 708 709 710
	if (oom_adjust < task->oomkilladj && !capable(CAP_SYS_RESOURCE)) {
		put_task_struct(task);
		return -EACCES;
	}
L
Linus Torvalds 已提交
711
	task->oomkilladj = oom_adjust;
712
	put_task_struct(task);
L
Linus Torvalds 已提交
713 714 715 716 717
	if (end - buffer == 0)
		return -EIO;
	return end - buffer;
}

718
static const struct file_operations proc_oom_adjust_operations = {
L
Linus Torvalds 已提交
719 720 721 722
	.read		= oom_adjust_read,
	.write		= oom_adjust_write,
};

723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756
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,
};

L
Linus Torvalds 已提交
757 758 759 760 761
#ifdef CONFIG_AUDITSYSCALL
#define TMPBUFLEN 21
static ssize_t proc_loginuid_read(struct file * file, char __user * buf,
				  size_t count, loff_t *ppos)
{
762
	struct inode * inode = file->f_path.dentry->d_inode;
763
	struct task_struct *task = get_proc_task(inode);
L
Linus Torvalds 已提交
764 765 766
	ssize_t length;
	char tmpbuf[TMPBUFLEN];

767 768
	if (!task)
		return -ESRCH;
L
Linus Torvalds 已提交
769 770
	length = scnprintf(tmpbuf, TMPBUFLEN, "%u",
				audit_get_loginuid(task->audit_context));
771
	put_task_struct(task);
L
Linus Torvalds 已提交
772 773 774 775 776 777
	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)
{
778
	struct inode * inode = file->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
779 780 781 782 783 784 785
	char *page, *tmp;
	ssize_t length;
	uid_t loginuid;

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

786
	if (current != pid_task(proc_pid(inode), PIDTYPE_PID))
L
Linus Torvalds 已提交
787 788
		return -EPERM;

789 790
	if (count >= PAGE_SIZE)
		count = PAGE_SIZE - 1;
L
Linus Torvalds 已提交
791 792 793 794 795 796 797 798 799 800 801 802

	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;

803
	page[count] = '\0';
L
Linus Torvalds 已提交
804 805 806 807 808 809
	loginuid = simple_strtoul(page, &tmp, 10);
	if (tmp == page) {
		length = -EINVAL;
		goto out_free_page;

	}
810
	length = audit_set_loginuid(current, loginuid);
L
Linus Torvalds 已提交
811 812 813 814 815 816 817 818
	if (likely(length == 0))
		length = count;

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

819
static const struct file_operations proc_loginuid_operations = {
L
Linus Torvalds 已提交
820 821 822 823 824 825 826 827 828
	.read		= proc_loginuid_read,
	.write		= proc_loginuid_write,
};
#endif

#ifdef CONFIG_SECCOMP
static ssize_t seccomp_read(struct file *file, char __user *buf,
			    size_t count, loff_t *ppos)
{
829
	struct task_struct *tsk = get_proc_task(file->f_dentry->d_inode);
L
Linus Torvalds 已提交
830 831 832 833
	char __buf[20];
	loff_t __ppos = *ppos;
	size_t len;

834 835
	if (!tsk)
		return -ESRCH;
L
Linus Torvalds 已提交
836 837
	/* no need to print the trailing zero, so use only len */
	len = sprintf(__buf, "%u\n", tsk->seccomp.mode);
838
	put_task_struct(tsk);
L
Linus Torvalds 已提交
839 840 841 842 843 844 845 846 847 848 849 850 851
	if (__ppos >= len)
		return 0;
	if (count > len - __ppos)
		count = len - __ppos;
	if (copy_to_user(buf, __buf + __ppos, count))
		return -EFAULT;
	*ppos = __ppos + count;
	return count;
}

static ssize_t seccomp_write(struct file *file, const char __user *buf,
			     size_t count, loff_t *ppos)
{
852
	struct task_struct *tsk = get_proc_task(file->f_dentry->d_inode);
L
Linus Torvalds 已提交
853 854
	char __buf[20], *end;
	unsigned int seccomp_mode;
855 856 857 858 859
	ssize_t result;

	result = -ESRCH;
	if (!tsk)
		goto out_no_task;
L
Linus Torvalds 已提交
860 861

	/* can set it only once to be even more secure */
862
	result = -EPERM;
L
Linus Torvalds 已提交
863
	if (unlikely(tsk->seccomp.mode))
864
		goto out;
L
Linus Torvalds 已提交
865

866
	result = -EFAULT;
L
Linus Torvalds 已提交
867 868 869
	memset(__buf, 0, sizeof(__buf));
	count = min(count, sizeof(__buf) - 1);
	if (copy_from_user(__buf, buf, count))
870 871
		goto out;

L
Linus Torvalds 已提交
872 873 874
	seccomp_mode = simple_strtoul(__buf, &end, 0);
	if (*end == '\n')
		end++;
875
	result = -EINVAL;
L
Linus Torvalds 已提交
876 877 878 879
	if (seccomp_mode && seccomp_mode <= NR_SECCOMP_MODES) {
		tsk->seccomp.mode = seccomp_mode;
		set_tsk_thread_flag(tsk, TIF_SECCOMP);
	} else
880 881
		goto out;
	result = -EIO;
L
Linus Torvalds 已提交
882
	if (unlikely(!(end - __buf)))
883 884 885 886 887 888
		goto out;
	result = end - __buf;
out:
	put_task_struct(tsk);
out_no_task:
	return result;
L
Linus Torvalds 已提交
889 890
}

891
static const struct file_operations proc_seccomp_operations = {
L
Linus Torvalds 已提交
892 893 894 895 896
	.read		= seccomp_read,
	.write		= seccomp_write,
};
#endif /* CONFIG_SECCOMP */

897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949
#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;
	loff_t __ppos = *ppos;

	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);
	if (__ppos >= len)
		return 0;
	if (count > len-__ppos)
		count = len-__ppos;
	if (copy_to_user(buf, buffer + __ppos, count))
		return -EFAULT;
	*ppos = __ppos + count;
	return count;
}

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;
}

950
static const struct file_operations proc_fault_inject_operations = {
951 952 953 954 955
	.read		= proc_fault_inject_read,
	.write		= proc_fault_inject_write,
};
#endif

956
static void *proc_pid_follow_link(struct dentry *dentry, struct nameidata *nd)
L
Linus Torvalds 已提交
957 958 959 960 961 962 963
{
	struct inode *inode = dentry->d_inode;
	int error = -EACCES;

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

964 965
	/* Are we allowed to snoop on the tasks file descriptors? */
	if (!proc_fd_access_allowed(inode))
L
Linus Torvalds 已提交
966 967 968 969 970
		goto out;

	error = PROC_I(inode)->op.proc_get_link(inode, &nd->dentry, &nd->mnt);
	nd->last_type = LAST_BIND;
out:
971
	return ERR_PTR(error);
L
Linus Torvalds 已提交
972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006
}

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;
		
	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;

1007 1008
	/* Are we allowed to snoop on the tasks file descriptors? */
	if (!proc_fd_access_allowed(inode))
L
Linus Torvalds 已提交
1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021
		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;
}

1022
static const struct inode_operations proc_pid_link_inode_operations = {
L
Linus Torvalds 已提交
1023
	.readlink	= proc_pid_readlink,
1024 1025
	.follow_link	= proc_pid_follow_link,
	.setattr	= proc_setattr,
L
Linus Torvalds 已提交
1026 1027
};

1028 1029 1030 1031

/* building an inode */

static int task_dumpable(struct task_struct *task)
L
Linus Torvalds 已提交
1032
{
1033 1034
	int dumpable = 0;
	struct mm_struct *mm;
L
Linus Torvalds 已提交
1035

1036 1037 1038 1039 1040 1041 1042 1043 1044
	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
1045 1046


1047
static struct inode *proc_pid_make_inode(struct super_block * sb, struct task_struct *task)
1048 1049 1050
{
	struct inode * inode;
	struct proc_inode *ei;
L
Linus Torvalds 已提交
1051

1052
	/* We need a new inode */
L
Linus Torvalds 已提交
1053

1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065
	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.
	 */
1066
	ei->pid = get_task_pid(task, PIDTYPE_PID);
1067 1068 1069 1070 1071 1072 1073 1074
	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
1075
	}
1076 1077
	security_task_to_inode(task, inode);

L
Linus Torvalds 已提交
1078
out:
1079 1080 1081 1082 1083
	return inode;

out_unlock:
	iput(inode);
	return NULL;
L
Linus Torvalds 已提交
1084 1085
}

1086
static int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
L
Linus Torvalds 已提交
1087 1088
{
	struct inode *inode = dentry->d_inode;
1089 1090
	struct task_struct *task;
	generic_fillattr(inode, stat);
L
Linus Torvalds 已提交
1091

1092 1093 1094 1095 1096 1097 1098 1099 1100
	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
1101 1102
		}
	}
1103
	rcu_read_unlock();
A
Alan Cox 已提交
1104
	return 0;
L
Linus Torvalds 已提交
1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115
}

/* 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.
1116 1117 1118 1119 1120 1121 1122
 *
 * 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
1123 1124 1125 1126
 */
static int pid_revalidate(struct dentry *dentry, struct nameidata *nd)
{
	struct inode *inode = dentry->d_inode;
1127 1128 1129 1130
	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
1131 1132 1133 1134 1135 1136
			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
1137
		inode->i_mode &= ~(S_ISUID | S_ISGID);
L
Linus Torvalds 已提交
1138
		security_task_to_inode(task, inode);
1139
		put_task_struct(task);
L
Linus Torvalds 已提交
1140 1141 1142 1143 1144 1145
		return 1;
	}
	d_drop(dentry);
	return 0;
}

1146
static int pid_delete_dentry(struct dentry * dentry)
1147
{
1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162
	/* 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 */

1163 1164
typedef struct dentry *instantiate_t(struct inode *, struct dentry *,
				struct task_struct *, const void *);
1165

1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177
/*
 * 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.
 */
1178 1179
static int proc_fill_cache(struct file *filp, void *dirent, filldir_t filldir,
	char *name, int len,
1180
	instantiate_t instantiate, struct task_struct *task, const void *ptr)
1181
{
1182
	struct dentry *child, *dir = filp->f_path.dentry;
1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219
	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);
}

1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241
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
1242 1243 1244 1245
#define PROC_FDINFO_MAX 64

static int proc_fd_info(struct inode *inode, struct dentry **dentry,
			struct vfsmount **mnt, char *info)
1246 1247 1248 1249 1250
{
	struct task_struct *task = get_proc_task(inode);
	struct files_struct *files = NULL;
	struct file *file;
	int fd = proc_fd(inode);
1251 1252

	if (task) {
1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263
		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
1264 1265 1266 1267 1268 1269 1270 1271 1272 1273
			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);
1274 1275 1276
			spin_unlock(&files->file_lock);
			put_files_struct(files);
			return 0;
1277
		}
1278 1279
		spin_unlock(&files->file_lock);
		put_files_struct(files);
1280
	}
1281
	return -ENOENT;
1282 1283
}

M
Miklos Szeredi 已提交
1284 1285 1286 1287 1288 1289
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
1290 1291 1292
static int tid_fd_revalidate(struct dentry *dentry, struct nameidata *nd)
{
	struct inode *inode = dentry->d_inode;
1293
	struct task_struct *task = get_proc_task(inode);
1294
	int fd = proc_fd(inode);
L
Linus Torvalds 已提交
1295 1296
	struct files_struct *files;

1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310
	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
1311
				inode->i_mode &= ~(S_ISUID | S_ISGID);
1312 1313 1314 1315
				security_task_to_inode(task, inode);
				put_task_struct(task);
				return 1;
			}
1316
			rcu_read_unlock();
L
Linus Torvalds 已提交
1317 1318
			put_files_struct(files);
		}
1319
		put_task_struct(task);
L
Linus Torvalds 已提交
1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330
	}
	d_drop(dentry);
	return 0;
}

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

1331
static struct dentry *proc_fd_instantiate(struct inode *dir,
1332
	struct dentry *dentry, struct task_struct *task, const void *ptr)
L
Linus Torvalds 已提交
1333
{
1334
	unsigned fd = *(const unsigned *)ptr;
1335 1336 1337 1338 1339
	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
1340

1341
	inode = proc_pid_make_inode(dir->i_sb, task);
L
Linus Torvalds 已提交
1342 1343 1344
	if (!inode)
		goto out;
	ei = PROC_I(inode);
1345
	ei->fd = fd;
L
Linus Torvalds 已提交
1346 1347
	files = get_files_struct(task);
	if (!files)
1348
		goto out_iput;
L
Linus Torvalds 已提交
1349
	inode->i_mode = S_IFLNK;
1350 1351 1352 1353 1354 1355

	/*
	 * 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
1356 1357
	file = fcheck_files(files, fd);
	if (!file)
1358
		goto out_unlock;
L
Linus Torvalds 已提交
1359 1360 1361 1362
	if (file->f_mode & 1)
		inode->i_mode |= S_IRUSR | S_IXUSR;
	if (file->f_mode & 2)
		inode->i_mode |= S_IWUSR | S_IXUSR;
1363
	spin_unlock(&files->file_lock);
L
Linus Torvalds 已提交
1364
	put_files_struct(files);
1365

L
Linus Torvalds 已提交
1366 1367 1368 1369 1370
	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);
1371 1372
	/* Close the race of the process dying before we return the dentry */
	if (tid_fd_revalidate(dentry, NULL))
1373
		error = NULL;
L
Linus Torvalds 已提交
1374

1375 1376 1377
 out:
	return error;
out_unlock:
1378
	spin_unlock(&files->file_lock);
L
Linus Torvalds 已提交
1379
	put_files_struct(files);
1380
out_iput:
L
Linus Torvalds 已提交
1381
	iput(inode);
1382
	goto out;
L
Linus Torvalds 已提交
1383 1384
}

M
Miklos Szeredi 已提交
1385 1386 1387
static struct dentry *proc_lookupfd_common(struct inode *dir,
					   struct dentry *dentry,
					   instantiate_t instantiate)
1388 1389 1390 1391 1392 1393 1394 1395 1396 1397
{
	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
1398
	result = instantiate(dir, dentry, task, &fd);
1399 1400 1401 1402 1403 1404
out:
	put_task_struct(task);
out_no_task:
	return result;
}

M
Miklos Szeredi 已提交
1405 1406
static int proc_readfd_common(struct file * filp, void * dirent,
			      filldir_t filldir, instantiate_t instantiate)
1407
{
1408
	struct dentry *dentry = filp->f_path.dentry;
1409 1410 1411 1412 1413 1414
	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
1415

1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441
	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
1442 1443
				char name[PROC_NUMBUF];
				int len;
1444 1445 1446 1447 1448

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

M
Miklos Szeredi 已提交
1449 1450 1451 1452
				len = snprintf(name, sizeof(name), "%d", fd);
				if (proc_fill_cache(filp, dirent, filldir,
						    name, len, instantiate,
						    p, &fd) < 0) {
1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466
					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
1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492
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,
};

1493
static const struct file_operations proc_fd_operations = {
1494 1495
	.read		= generic_read_dir,
	.readdir	= proc_readfd,
L
Linus Torvalds 已提交
1496 1497
};

1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514
/*
 * /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
1515 1516 1517
/*
 * proc directories can do almost nothing..
 */
1518
static const struct inode_operations proc_fd_inode_operations = {
L
Linus Torvalds 已提交
1519
	.lookup		= proc_lookupfd,
1520
	.permission	= proc_fd_permission,
1521
	.setattr	= proc_setattr,
L
Linus Torvalds 已提交
1522 1523
};

M
Miklos Szeredi 已提交
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 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575
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,
};


1576
static struct dentry *proc_pident_instantiate(struct inode *dir,
1577
	struct dentry *dentry, struct task_struct *task, const void *ptr)
1578
{
1579
	const struct pid_entry *p = ptr;
1580 1581 1582 1583
	struct inode *inode;
	struct proc_inode *ei;
	struct dentry *error = ERR_PTR(-EINVAL);

1584
	inode = proc_pid_make_inode(dir->i_sb, task);
1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605
	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
1606 1607
static struct dentry *proc_pident_lookup(struct inode *dir, 
					 struct dentry *dentry,
1608
					 const struct pid_entry *ents,
1609
					 unsigned int nents)
L
Linus Torvalds 已提交
1610 1611
{
	struct inode *inode;
1612
	struct dentry *error;
1613
	struct task_struct *task = get_proc_task(dir);
1614
	const struct pid_entry *p, *last;
L
Linus Torvalds 已提交
1615

1616
	error = ERR_PTR(-ENOENT);
L
Linus Torvalds 已提交
1617 1618
	inode = NULL;

1619 1620
	if (!task)
		goto out_no_task;
L
Linus Torvalds 已提交
1621

1622 1623 1624 1625
	/*
	 * Yes, it does not scale. And it should not. Don't add
	 * new entries into /proc/<tgid>/ without very good reasons.
	 */
1626 1627
	last = &ents[nents - 1];
	for (p = ents; p <= last; p++) {
L
Linus Torvalds 已提交
1628 1629 1630 1631 1632
		if (p->len != dentry->d_name.len)
			continue;
		if (!memcmp(dentry->d_name.name, p->name, p->len))
			break;
	}
1633
	if (p > last)
L
Linus Torvalds 已提交
1634 1635
		goto out;

1636
	error = proc_pident_instantiate(dir, dentry, task, p);
L
Linus Torvalds 已提交
1637
out:
1638 1639
	put_task_struct(task);
out_no_task:
1640
	return error;
L
Linus Torvalds 已提交
1641 1642
}

1643 1644
static int proc_pident_fill_cache(struct file *filp, void *dirent,
	filldir_t filldir, struct task_struct *task, const struct pid_entry *p)
1645 1646 1647 1648 1649
{
	return proc_fill_cache(filp, dirent, filldir, p->name, p->len,
				proc_pident_instantiate, task, p);
}

1650 1651
static int proc_pident_readdir(struct file *filp,
		void *dirent, filldir_t filldir,
1652
		const struct pid_entry *ents, unsigned int nents)
1653 1654 1655
{
	int i;
	int pid;
1656
	struct dentry *dentry = filp->f_path.dentry;
1657 1658
	struct inode *inode = dentry->d_inode;
	struct task_struct *task = get_proc_task(inode);
1659
	const struct pid_entry *p, *last;
1660 1661 1662 1663 1664
	ino_t ino;
	int ret;

	ret = -ENOENT;
	if (!task)
1665
		goto out_no_task;
1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691

	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;
1692 1693
		last = &ents[nents - 1];
		while (p <= last) {
1694
			if (proc_pident_fill_cache(filp, dirent, filldir, task, p) < 0)
1695 1696 1697 1698 1699 1700 1701 1702
				goto out;
			filp->f_pos++;
			p++;
		}
	}

	ret = 1;
out:
1703 1704
	put_task_struct(task);
out_no_task:
1705
	return ret;
L
Linus Torvalds 已提交
1706 1707
}

1708 1709 1710 1711
#ifdef CONFIG_SECURITY
static ssize_t proc_pid_attr_read(struct file * file, char __user * buf,
				  size_t count, loff_t *ppos)
{
1712
	struct inode * inode = file->f_path.dentry->d_inode;
1713
	char *p = NULL;
1714 1715 1716 1717
	ssize_t length;
	struct task_struct *task = get_proc_task(inode);

	if (!task)
1718
		return -ESRCH;
1719 1720

	length = security_getprocattr(task,
1721
				      (char*)file->f_path.dentry->d_name.name,
1722
				      &p);
1723
	put_task_struct(task);
1724 1725 1726
	if (length > 0)
		length = simple_read_from_buffer(buf, count, ppos, p, length);
	kfree(p);
1727
	return length;
L
Linus Torvalds 已提交
1728 1729
}

1730 1731 1732
static ssize_t proc_pid_attr_write(struct file * file, const char __user * buf,
				   size_t count, loff_t *ppos)
{
1733
	struct inode * inode = file->f_path.dentry->d_inode;
1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758
	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,
1759
				      (char*)file->f_path.dentry->d_name.name,
1760 1761 1762 1763 1764 1765 1766 1767 1768
				      (void*)page, count);
out_free:
	free_page((unsigned long) page);
out:
	put_task_struct(task);
out_no_task:
	return length;
}

1769
static const struct file_operations proc_pid_attr_operations = {
1770 1771 1772 1773
	.read		= proc_pid_attr_read,
	.write		= proc_pid_attr_write,
};

1774
static const struct pid_entry attr_dir_stuff[] = {
1775 1776 1777 1778 1779 1780
	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),
1781 1782
};

1783
static int proc_attr_dir_readdir(struct file * filp,
1784 1785 1786
			     void * dirent, filldir_t filldir)
{
	return proc_pident_readdir(filp,dirent,filldir,
1787
				   attr_dir_stuff,ARRAY_SIZE(attr_dir_stuff));
1788 1789
}

1790
static const struct file_operations proc_attr_dir_operations = {
L
Linus Torvalds 已提交
1791
	.read		= generic_read_dir,
1792
	.readdir	= proc_attr_dir_readdir,
L
Linus Torvalds 已提交
1793 1794
};

1795
static struct dentry *proc_attr_dir_lookup(struct inode *dir,
1796 1797
				struct dentry *dentry, struct nameidata *nd)
{
1798 1799
	return proc_pident_lookup(dir, dentry,
				  attr_dir_stuff, ARRAY_SIZE(attr_dir_stuff));
1800 1801
}

1802
static const struct inode_operations proc_attr_dir_inode_operations = {
1803
	.lookup		= proc_attr_dir_lookup,
1804
	.getattr	= pid_getattr,
1805
	.setattr	= proc_setattr,
L
Linus Torvalds 已提交
1806 1807
};

1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827
#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));
}

1828
static const struct inode_operations proc_self_inode_operations = {
1829 1830 1831 1832
	.readlink	= proc_self_readlink,
	.follow_link	= proc_self_follow_link,
};

1833 1834 1835 1836 1837 1838 1839
/*
 * 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.
 */
1840
static const struct pid_entry proc_base_stuff[] = {
1841
	NOD("self", S_IFLNK|S_IRWXUGO,
1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867
		&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,
};

1868
static struct dentry *proc_base_instantiate(struct inode *dir,
1869
	struct dentry *dentry, struct task_struct *task, const void *ptr)
1870
{
1871
	const struct pid_entry *p = ptr;
1872 1873
	struct inode *inode;
	struct proc_inode *ei;
1874
	struct dentry *error = ERR_PTR(-EINVAL);
1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888

	/* 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.
	 */
1889
	ei->pid = get_task_pid(task, PIDTYPE_PID);
1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914
	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;
}

1915 1916 1917 1918
static struct dentry *proc_base_lookup(struct inode *dir, struct dentry *dentry)
{
	struct dentry *error;
	struct task_struct *task = get_proc_task(dir);
1919
	const struct pid_entry *p, *last;
1920 1921 1922 1923 1924 1925 1926

	error = ERR_PTR(-ENOENT);

	if (!task)
		goto out_no_task;

	/* Lookup the directory entry */
1927 1928
	last = &proc_base_stuff[ARRAY_SIZE(proc_base_stuff) - 1];
	for (p = proc_base_stuff; p <= last; p++) {
1929 1930 1931 1932 1933
		if (p->len != dentry->d_name.len)
			continue;
		if (!memcmp(dentry->d_name.name, p->name, p->len))
			break;
	}
1934
	if (p > last)
1935 1936 1937 1938 1939 1940 1941 1942 1943 1944
		goto out;

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

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

1945 1946
static int proc_base_fill_cache(struct file *filp, void *dirent,
	filldir_t filldir, struct task_struct *task, const struct pid_entry *p)
1947 1948 1949 1950 1951
{
	return proc_fill_cache(filp, dirent, filldir, p->name, p->len,
				proc_base_instantiate, task, p);
}

1952 1953 1954 1955
#ifdef CONFIG_TASK_IO_ACCOUNTING
static int proc_pid_io_accounting(struct task_struct *task, char *buffer)
{
	return sprintf(buffer,
1956
#ifdef CONFIG_TASK_XACCT
1957 1958 1959 1960
			"rchar: %llu\n"
			"wchar: %llu\n"
			"syscr: %llu\n"
			"syscw: %llu\n"
1961
#endif
1962 1963 1964
			"read_bytes: %llu\n"
			"write_bytes: %llu\n"
			"cancelled_write_bytes: %llu\n",
1965
#ifdef CONFIG_TASK_XACCT
1966 1967 1968 1969
			(unsigned long long)task->rchar,
			(unsigned long long)task->wchar,
			(unsigned long long)task->syscr,
			(unsigned long long)task->syscw,
1970
#endif
1971 1972 1973 1974 1975 1976
			(unsigned long long)task->ioac.read_bytes,
			(unsigned long long)task->ioac.write_bytes,
			(unsigned long long)task->ioac.cancelled_write_bytes);
}
#endif

1977 1978 1979
/*
 * Thread groups
 */
1980
static const struct file_operations proc_task_operations;
1981
static const struct inode_operations proc_task_inode_operations;
1982

1983
static const struct pid_entry tgid_base_stuff[] = {
1984 1985
	DIR("task",       S_IRUGO|S_IXUGO, task),
	DIR("fd",         S_IRUSR|S_IXUSR, fd),
M
Miklos Szeredi 已提交
1986
	DIR("fdinfo",     S_IRUSR|S_IXUSR, fdinfo),
1987 1988 1989 1990 1991 1992 1993
	INF("environ",    S_IRUSR, pid_environ),
	INF("auxv",       S_IRUSR, pid_auxv),
	INF("status",     S_IRUGO, pid_status),
	INF("cmdline",    S_IRUGO, pid_cmdline),
	INF("stat",       S_IRUGO, tgid_stat),
	INF("statm",      S_IRUGO, pid_statm),
	REG("maps",       S_IRUGO, maps),
1994
#ifdef CONFIG_NUMA
1995
	REG("numa_maps",  S_IRUGO, numa_maps),
1996
#endif
1997
	REG("mem",        S_IRUSR|S_IWUSR, mem),
1998
#ifdef CONFIG_SECCOMP
1999
	REG("seccomp",    S_IRUSR|S_IWUSR, seccomp),
2000
#endif
2001 2002 2003 2004 2005
	LNK("cwd",        cwd),
	LNK("root",       root),
	LNK("exe",        exe),
	REG("mounts",     S_IRUGO, mounts),
	REG("mountstats", S_IRUSR, mountstats),
2006
#ifdef CONFIG_MMU
2007
	REG("clear_refs", S_IWUSR, clear_refs),
2008
	REG("smaps",      S_IRUGO, smaps),
2009 2010
#endif
#ifdef CONFIG_SECURITY
2011
	DIR("attr",       S_IRUGO|S_IXUGO, attr_dir),
2012 2013
#endif
#ifdef CONFIG_KALLSYMS
2014
	INF("wchan",      S_IRUGO, pid_wchan),
2015 2016
#endif
#ifdef CONFIG_SCHEDSTATS
2017
	INF("schedstat",  S_IRUGO, pid_schedstat),
2018 2019
#endif
#ifdef CONFIG_CPUSETS
2020
	REG("cpuset",     S_IRUGO, cpuset),
2021
#endif
2022 2023
	INF("oom_score",  S_IRUGO, oom_score),
	REG("oom_adj",    S_IRUGO|S_IWUSR, oom_adjust),
2024
#ifdef CONFIG_AUDITSYSCALL
2025
	REG("loginuid",   S_IWUSR|S_IRUGO, loginuid),
2026
#endif
2027 2028 2029
#ifdef CONFIG_FAULT_INJECTION
	REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject),
#endif
2030 2031 2032
#ifdef CONFIG_TASK_IO_ACCOUNTING
	INF("io",	S_IRUGO, pid_io_accounting),
#endif
2033
};
L
Linus Torvalds 已提交
2034

2035
static int proc_tgid_base_readdir(struct file * filp,
L
Linus Torvalds 已提交
2036 2037 2038
			     void * dirent, filldir_t filldir)
{
	return proc_pident_readdir(filp,dirent,filldir,
2039
				   tgid_base_stuff,ARRAY_SIZE(tgid_base_stuff));
L
Linus Torvalds 已提交
2040 2041
}

2042
static const struct file_operations proc_tgid_base_operations = {
L
Linus Torvalds 已提交
2043
	.read		= generic_read_dir,
2044
	.readdir	= proc_tgid_base_readdir,
L
Linus Torvalds 已提交
2045 2046
};

2047
static struct dentry *proc_tgid_base_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd){
2048 2049
	return proc_pident_lookup(dir, dentry,
				  tgid_base_stuff, ARRAY_SIZE(tgid_base_stuff));
L
Linus Torvalds 已提交
2050 2051
}

2052
static const struct inode_operations proc_tgid_base_inode_operations = {
2053
	.lookup		= proc_tgid_base_lookup,
2054
	.getattr	= pid_getattr,
2055
	.setattr	= proc_setattr,
L
Linus Torvalds 已提交
2056 2057 2058
};

/**
2059 2060 2061 2062 2063 2064 2065 2066 2067
 * 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
2068
 *
2069 2070 2071 2072 2073
 * 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
2074
 *
2075 2076 2077
 * 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
2078
 */
2079
void proc_flush_task(struct task_struct *task)
L
Linus Torvalds 已提交
2080
{
2081
	struct dentry *dentry, *leader, *dir;
2082
	char buf[PROC_NUMBUF];
2083 2084 2085 2086 2087 2088 2089 2090 2091 2092
	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
2093

2094 2095
	if (thread_group_leader(task))
		goto out;
L
Linus Torvalds 已提交
2096

2097 2098 2099 2100 2101
	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
2102

2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115
	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
2116
	}
2117 2118 2119 2120 2121 2122

	dput(dir);
out_put_leader:
	dput(leader);
out:
	return;
L
Linus Torvalds 已提交
2123 2124
}

2125 2126
static struct dentry *proc_pid_instantiate(struct inode *dir,
					   struct dentry * dentry,
2127
					   struct task_struct *task, const void *ptr)
2128 2129 2130 2131
{
	struct dentry *error = ERR_PTR(-ENOENT);
	struct inode *inode;

2132
	inode = proc_pid_make_inode(dir->i_sb, task);
2133 2134 2135 2136 2137 2138 2139
	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
2140
	inode->i_nlink = 5;
2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154
#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
2155 2156
struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *nd)
{
2157
	struct dentry *result = ERR_PTR(-ENOENT);
L
Linus Torvalds 已提交
2158 2159 2160
	struct task_struct *task;
	unsigned tgid;

2161 2162 2163 2164
	result = proc_base_lookup(dir, dentry);
	if (!IS_ERR(result) || PTR_ERR(result) != -ENOENT)
		goto out;

L
Linus Torvalds 已提交
2165 2166 2167 2168
	tgid = name_to_int(dentry);
	if (tgid == ~0U)
		goto out;

2169
	rcu_read_lock();
L
Linus Torvalds 已提交
2170 2171 2172
	task = find_task_by_pid(tgid);
	if (task)
		get_task_struct(task);
2173
	rcu_read_unlock();
L
Linus Torvalds 已提交
2174 2175 2176
	if (!task)
		goto out;

2177
	result = proc_pid_instantiate(dir, dentry, task, NULL);
L
Linus Torvalds 已提交
2178 2179
	put_task_struct(task);
out:
2180
	return result;
L
Linus Torvalds 已提交
2181 2182 2183
}

/*
2184
 * Find the first task with tgid >= tgid
2185
 *
L
Linus Torvalds 已提交
2186
 */
2187
static struct task_struct *next_tgid(unsigned int tgid)
L
Linus Torvalds 已提交
2188
{
2189 2190
	struct task_struct *task;
	struct pid *pid;
L
Linus Torvalds 已提交
2191

2192
	rcu_read_lock();
2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213
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);
2214
	}
2215
	rcu_read_unlock();
2216
	return task;
L
Linus Torvalds 已提交
2217 2218
}

2219
#define TGID_OFFSET (FIRST_PROCESS_ENTRY + ARRAY_SIZE(proc_base_stuff))
2220

2221 2222 2223 2224 2225 2226 2227 2228 2229
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
2230 2231 2232 2233
/* 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;
2234
	struct task_struct *reaper = get_proc_task(filp->f_path.dentry->d_inode);
2235 2236
	struct task_struct *task;
	int tgid;
L
Linus Torvalds 已提交
2237

2238 2239 2240
	if (!reaper)
		goto out_no_task;

2241
	for (; nr < ARRAY_SIZE(proc_base_stuff); filp->f_pos++, nr++) {
2242
		const struct pid_entry *p = &proc_base_stuff[nr];
2243
		if (proc_base_fill_cache(filp, dirent, filldir, reaper, p) < 0)
2244
			goto out;
L
Linus Torvalds 已提交
2245 2246
	}

2247 2248
	tgid = filp->f_pos - TGID_OFFSET;
	for (task = next_tgid(tgid);
2249
	     task;
2250
	     put_task_struct(task), task = next_tgid(tgid + 1)) {
2251
		tgid = task->pid;
2252
		filp->f_pos = tgid + TGID_OFFSET;
2253
		if (proc_pid_fill_cache(filp, dirent, filldir, task, tgid) < 0) {
2254
			put_task_struct(task);
2255
			goto out;
L
Linus Torvalds 已提交
2256
		}
2257
	}
2258 2259
	filp->f_pos = PID_MAX_LIMIT + TGID_OFFSET;
out:
2260 2261
	put_task_struct(reaper);
out_no_task:
2262 2263
	return 0;
}
L
Linus Torvalds 已提交
2264

2265 2266 2267
/*
 * Tasks
 */
2268
static const struct pid_entry tid_base_stuff[] = {
2269
	DIR("fd",        S_IRUSR|S_IXUSR, fd),
M
Miklos Szeredi 已提交
2270
	DIR("fdinfo",    S_IRUSR|S_IXUSR, fdinfo),
2271 2272 2273 2274 2275 2276 2277
	INF("environ",   S_IRUSR, pid_environ),
	INF("auxv",      S_IRUSR, pid_auxv),
	INF("status",    S_IRUGO, pid_status),
	INF("cmdline",   S_IRUGO, pid_cmdline),
	INF("stat",      S_IRUGO, tid_stat),
	INF("statm",     S_IRUGO, pid_statm),
	REG("maps",      S_IRUGO, maps),
2278
#ifdef CONFIG_NUMA
2279
	REG("numa_maps", S_IRUGO, numa_maps),
2280
#endif
2281
	REG("mem",       S_IRUSR|S_IWUSR, mem),
2282
#ifdef CONFIG_SECCOMP
2283
	REG("seccomp",   S_IRUSR|S_IWUSR, seccomp),
2284
#endif
2285 2286 2287 2288
	LNK("cwd",       cwd),
	LNK("root",      root),
	LNK("exe",       exe),
	REG("mounts",    S_IRUGO, mounts),
2289
#ifdef CONFIG_MMU
2290
	REG("clear_refs", S_IWUSR, clear_refs),
2291
	REG("smaps",     S_IRUGO, smaps),
2292 2293
#endif
#ifdef CONFIG_SECURITY
2294
	DIR("attr",      S_IRUGO|S_IXUGO, attr_dir),
2295 2296
#endif
#ifdef CONFIG_KALLSYMS
2297
	INF("wchan",     S_IRUGO, pid_wchan),
2298 2299
#endif
#ifdef CONFIG_SCHEDSTATS
2300
	INF("schedstat", S_IRUGO, pid_schedstat),
2301 2302
#endif
#ifdef CONFIG_CPUSETS
2303
	REG("cpuset",    S_IRUGO, cpuset),
2304
#endif
2305 2306
	INF("oom_score", S_IRUGO, oom_score),
	REG("oom_adj",   S_IRUGO|S_IWUSR, oom_adjust),
2307
#ifdef CONFIG_AUDITSYSCALL
2308
	REG("loginuid",  S_IWUSR|S_IRUGO, loginuid),
2309
#endif
2310 2311 2312
#ifdef CONFIG_FAULT_INJECTION
	REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject),
#endif
2313 2314 2315 2316 2317 2318 2319 2320 2321 2322
};

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){
2323 2324
	return proc_pident_lookup(dir, dentry,
				  tid_base_stuff, ARRAY_SIZE(tid_base_stuff));
2325 2326
}

2327
static const struct file_operations proc_tid_base_operations = {
2328 2329 2330 2331
	.read		= generic_read_dir,
	.readdir	= proc_tid_base_readdir,
};

2332
static const struct inode_operations proc_tid_base_inode_operations = {
2333 2334 2335 2336 2337
	.lookup		= proc_tid_base_lookup,
	.getattr	= pid_getattr,
	.setattr	= proc_setattr,
};

2338
static struct dentry *proc_task_instantiate(struct inode *dir,
2339
	struct dentry *dentry, struct task_struct *task, const void *ptr)
2340 2341 2342
{
	struct dentry *error = ERR_PTR(-ENOENT);
	struct inode *inode;
2343
	inode = proc_pid_make_inode(dir->i_sb, task);
2344 2345 2346 2347 2348 2349 2350

	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
2351
	inode->i_nlink = 4;
2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365
#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;
}

2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389
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;

2390
	result = proc_task_instantiate(dir, dentry, task, NULL);
2391 2392 2393 2394 2395 2396 2397 2398
out_drop_task:
	put_task_struct(task);
out:
	put_task_struct(leader);
out_no_task:
	return result;
}

2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410
/*
 * 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.
 */
2411 2412
static struct task_struct *first_tid(struct task_struct *leader,
					int tid, int nr)
2413
{
O
Oleg Nesterov 已提交
2414
	struct task_struct *pos;
L
Linus Torvalds 已提交
2415

2416
	rcu_read_lock();
2417 2418 2419
	/* 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
2420 2421
		if (pos && (pos->group_leader == leader))
			goto found;
2422
	}
L
Linus Torvalds 已提交
2423

2424
	/* If nr exceeds the number of threads there is nothing todo */
O
Oleg Nesterov 已提交
2425 2426 2427
	pos = NULL;
	if (nr && nr >= get_nr_threads(leader))
		goto out;
L
Linus Torvalds 已提交
2428

O
Oleg Nesterov 已提交
2429 2430
	/* If we haven't found our starting place yet start
	 * with the leader and walk nr threads forward.
2431
	 */
O
Oleg Nesterov 已提交
2432 2433 2434 2435 2436 2437
	for (pos = leader; nr > 0; --nr) {
		pos = next_thread(pos);
		if (pos == leader) {
			pos = NULL;
			goto out;
		}
L
Linus Torvalds 已提交
2438
	}
O
Oleg Nesterov 已提交
2439 2440 2441
found:
	get_task_struct(pos);
out:
2442
	rcu_read_unlock();
2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453
	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
2454
	struct task_struct *pos = NULL;
2455
	rcu_read_lock();
O
Oleg Nesterov 已提交
2456
	if (pid_alive(start)) {
2457
		pos = next_thread(start);
O
Oleg Nesterov 已提交
2458 2459 2460 2461 2462
		if (thread_group_leader(pos))
			pos = NULL;
		else
			get_task_struct(pos);
	}
2463
	rcu_read_unlock();
2464 2465
	put_task_struct(start);
	return pos;
L
Linus Torvalds 已提交
2466 2467
}

2468 2469 2470 2471 2472 2473 2474 2475 2476
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
2477 2478 2479
/* for the /proc/TGID/task/ directories */
static int proc_task_readdir(struct file * filp, void * dirent, filldir_t filldir)
{
2480
	struct dentry *dentry = filp->f_path.dentry;
L
Linus Torvalds 已提交
2481
	struct inode *inode = dentry->d_inode;
2482
	struct task_struct *leader = NULL;
2483
	struct task_struct *task;
L
Linus Torvalds 已提交
2484 2485
	int retval = -ENOENT;
	ino_t ino;
2486
	int tid;
L
Linus Torvalds 已提交
2487 2488
	unsigned long pos = filp->f_pos;  /* avoiding "long long" filp->f_pos */

2489 2490 2491 2492 2493 2494 2495 2496 2497 2498
	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);
2499 2500
	if (!leader)
		goto out_no_task;
L
Linus Torvalds 已提交
2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517
	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 */
	}

2518 2519 2520 2521 2522 2523 2524 2525 2526
	/* 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;
2527
		if (proc_task_fill_cache(filp, dirent, filldir, task, tid) < 0) {
2528 2529 2530 2531
			/* 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
2532
			break;
2533
		}
L
Linus Torvalds 已提交
2534 2535 2536
	}
out:
	filp->f_pos = pos;
2537 2538
	put_task_struct(leader);
out_no_task:
L
Linus Torvalds 已提交
2539 2540
	return retval;
}
2541 2542 2543 2544

static int proc_task_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
{
	struct inode *inode = dentry->d_inode;
2545
	struct task_struct *p = get_proc_task(inode);
2546 2547
	generic_fillattr(inode, stat);

2548 2549 2550 2551 2552
	if (p) {
		rcu_read_lock();
		stat->nlink += get_nr_threads(p);
		rcu_read_unlock();
		put_task_struct(p);
2553 2554 2555 2556
	}

	return 0;
}
2557

2558
static const struct inode_operations proc_task_inode_operations = {
2559 2560 2561 2562 2563
	.lookup		= proc_task_lookup,
	.getattr	= proc_task_getattr,
	.setattr	= proc_setattr,
};

2564
static const struct file_operations proc_task_operations = {
2565 2566 2567
	.read		= generic_read_dir,
	.readdir	= proc_task_readdir,
};