util.c 19.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9
/*
 * linux/ipc/util.c
 * Copyright (C) 1992 Krishna Balasubramanian
 *
 * Sep 1997 - Call suser() last after "normal" permission checks so we
 *            get BSD style process accounting right.
 *            Occurs in several places in the IPC code.
 *            Chris Evans, <chris@ferret.lmh.ox.ac.uk>
 * Nov 1999 - ipc helper functions, unified SMP locking
10
 *	      Manfred Spraul <manfred@colorfullife.com>
L
Linus Torvalds 已提交
11 12
 * Oct 2002 - One lock per IPC id. RCU ipc_free for lock-free grow_ary().
 *            Mingming Cao <cmm@us.ibm.com>
S
Steve Grubb 已提交
13 14
 * Mar 2006 - support for audit of ipc object properties
 *            Dustin Kirkland <dustin.kirkland@us.ibm.com>
K
Kirill Korotaev 已提交
15 16 17
 * Jun 2006 - namespaces ssupport
 *            OpenVZ, SWsoft Inc.
 *            Pavel Emelianov <xemul@openvz.org>
L
Linus Torvalds 已提交
18 19 20 21 22 23 24 25 26
 */

#include <linux/mm.h>
#include <linux/shm.h>
#include <linux/init.h>
#include <linux/msg.h>
#include <linux/smp_lock.h>
#include <linux/vmalloc.h>
#include <linux/slab.h>
27
#include <linux/capability.h>
L
Linus Torvalds 已提交
28 29 30 31
#include <linux/highuid.h>
#include <linux/security.h>
#include <linux/rcupdate.h>
#include <linux/workqueue.h>
32 33
#include <linux/seq_file.h>
#include <linux/proc_fs.h>
S
Steve Grubb 已提交
34
#include <linux/audit.h>
K
Kirill Korotaev 已提交
35
#include <linux/nsproxy.h>
L
Linus Torvalds 已提交
36 37 38 39 40

#include <asm/unistd.h>

#include "util.h"

41 42 43
struct ipc_proc_iface {
	const char *path;
	const char *header;
K
Kirill Korotaev 已提交
44
	int ids;
45 46 47
	int (*show)(struct seq_file *, void *);
};

K
Kirill Korotaev 已提交
48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146
struct ipc_namespace init_ipc_ns = {
	.kref = {
		.refcount	= ATOMIC_INIT(2),
	},
};

#ifdef CONFIG_IPC_NS
static struct ipc_namespace *clone_ipc_ns(struct ipc_namespace *old_ns)
{
	int err;
	struct ipc_namespace *ns;

	err = -ENOMEM;
	ns = kmalloc(sizeof(struct ipc_namespace), GFP_KERNEL);
	if (ns == NULL)
		goto err_mem;

	err = sem_init_ns(ns);
	if (err)
		goto err_sem;
	err = msg_init_ns(ns);
	if (err)
		goto err_msg;
	err = shm_init_ns(ns);
	if (err)
		goto err_shm;

	kref_init(&ns->kref);
	return ns;

err_shm:
	msg_exit_ns(ns);
err_msg:
	sem_exit_ns(ns);
err_sem:
	kfree(ns);
err_mem:
	return ERR_PTR(err);
}

int unshare_ipcs(unsigned long unshare_flags, struct ipc_namespace **new_ipc)
{
	struct ipc_namespace *new;

	if (unshare_flags & CLONE_NEWIPC) {
		if (!capable(CAP_SYS_ADMIN))
			return -EPERM;

		new = clone_ipc_ns(current->nsproxy->ipc_ns);
		if (IS_ERR(new))
			return PTR_ERR(new);

		*new_ipc = new;
	}

	return 0;
}

int copy_ipcs(unsigned long flags, struct task_struct *tsk)
{
	struct ipc_namespace *old_ns = tsk->nsproxy->ipc_ns;
	struct ipc_namespace *new_ns;
	int err = 0;

	if (!old_ns)
		return 0;

	get_ipc_ns(old_ns);

	if (!(flags & CLONE_NEWIPC))
		return 0;

	if (!capable(CAP_SYS_ADMIN)) {
		err = -EPERM;
		goto out;
	}

	new_ns = clone_ipc_ns(old_ns);
	if (!new_ns) {
		err = -ENOMEM;
		goto out;
	}

	tsk->nsproxy->ipc_ns = new_ns;
out:
	put_ipc_ns(old_ns);
	return err;
}

void free_ipc_ns(struct kref *kref)
{
	struct ipc_namespace *ns;

	ns = container_of(kref, struct ipc_namespace, kref);
	sem_exit_ns(ns);
	msg_exit_ns(ns);
	shm_exit_ns(ns);
	kfree(ns);
}
147 148 149 150 151 152 153
#else
int copy_ipcs(unsigned long flags, struct task_struct *tsk)
{
	if (flags & CLONE_NEWIPC)
		return -EINVAL;
	return 0;
}
K
Kirill Korotaev 已提交
154 155
#endif

L
Linus Torvalds 已提交
156 157 158 159
/**
 *	ipc_init	-	initialise IPC subsystem
 *
 *	The various system5 IPC resources (semaphores, messages and shared
160
 *	memory) are initialised
L
Linus Torvalds 已提交
161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181
 */
 
static int __init ipc_init(void)
{
	sem_init();
	msg_init();
	shm_init();
	return 0;
}
__initcall(ipc_init);

/**
 *	ipc_init_ids		-	initialise IPC identifiers
 *	@ids: Identifier set
 *	@size: Number of identifiers
 *
 *	Given a size for the ipc identifier range (limited below IPCMNI)
 *	set up the sequence range to use then allocate and initialise the
 *	array itself. 
 */
 
K
Kirill Korotaev 已提交
182
void __ipc_init ipc_init_ids(struct ipc_ids* ids, int size)
L
Linus Torvalds 已提交
183 184
{
	int i;
I
Ingo Molnar 已提交
185 186

	mutex_init(&ids->mutex);
L
Linus Torvalds 已提交
187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213

	if(size > IPCMNI)
		size = IPCMNI;
	ids->in_use = 0;
	ids->max_id = -1;
	ids->seq = 0;
	{
		int seq_limit = INT_MAX/SEQ_MULTIPLIER;
		if(seq_limit > USHRT_MAX)
			ids->seq_max = USHRT_MAX;
		 else
		 	ids->seq_max = seq_limit;
	}

	ids->entries = ipc_rcu_alloc(sizeof(struct kern_ipc_perm *)*size +
				     sizeof(struct ipc_id_ary));

	if(ids->entries == NULL) {
		printk(KERN_ERR "ipc_init_ids() failed, ipc service disabled.\n");
		size = 0;
		ids->entries = &ids->nullentry;
	}
	ids->entries->size = size;
	for(i=0;i<size;i++)
		ids->entries->p[i] = NULL;
}

214
#ifdef CONFIG_PROC_FS
215
static const struct file_operations sysvipc_proc_fops;
216
/**
217
 *	ipc_init_proc_interface	-  Create a proc interface for sysipc types using a seq_file interface.
218 219 220 221 222 223
 *	@path: Path in procfs
 *	@header: Banner to be printed at the beginning of the file.
 *	@ids: ipc id table to iterate.
 *	@show: show routine.
 */
void __init ipc_init_proc_interface(const char *path, const char *header,
K
Kirill Korotaev 已提交
224
		int ids, int (*show)(struct seq_file *, void *))
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248
{
	struct proc_dir_entry *pde;
	struct ipc_proc_iface *iface;

	iface = kmalloc(sizeof(*iface), GFP_KERNEL);
	if (!iface)
		return;
	iface->path	= path;
	iface->header	= header;
	iface->ids	= ids;
	iface->show	= show;

	pde = create_proc_entry(path,
				S_IRUGO,        /* world readable */
				NULL            /* parent dir */);
	if (pde) {
		pde->data = iface;
		pde->proc_fops = &sysvipc_proc_fops;
	} else {
		kfree(iface);
	}
}
#endif

L
Linus Torvalds 已提交
249 250 251 252 253
/**
 *	ipc_findkey	-	find a key in an ipc identifier set	
 *	@ids: Identifier set
 *	@key: The key to find
 *	
I
Ingo Molnar 已提交
254
 *	Requires ipc_ids.mutex locked.
L
Linus Torvalds 已提交
255 256 257 258 259 260 261 262 263 264 265
 *	Returns the identifier if found or -1 if not.
 */
 
int ipc_findkey(struct ipc_ids* ids, key_t key)
{
	int id;
	struct kern_ipc_perm* p;
	int max_id = ids->max_id;

	/*
	 * rcu_dereference() is not needed here
I
Ingo Molnar 已提交
266
	 * since ipc_ids.mutex is held
L
Linus Torvalds 已提交
267 268 269 270 271 272 273 274 275 276 277 278
	 */
	for (id = 0; id <= max_id; id++) {
		p = ids->entries->p[id];
		if(p==NULL)
			continue;
		if (key == p->key)
			return id;
	}
	return -1;
}

/*
I
Ingo Molnar 已提交
279
 * Requires ipc_ids.mutex locked
L
Linus Torvalds 已提交
280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297
 */
static int grow_ary(struct ipc_ids* ids, int newsize)
{
	struct ipc_id_ary* new;
	struct ipc_id_ary* old;
	int i;
	int size = ids->entries->size;

	if(newsize > IPCMNI)
		newsize = IPCMNI;
	if(newsize <= size)
		return newsize;

	new = ipc_rcu_alloc(sizeof(struct kern_ipc_perm *)*newsize +
			    sizeof(struct ipc_id_ary));
	if(new == NULL)
		return size;
	new->size = newsize;
298
	memcpy(new->p, ids->entries->p, sizeof(struct kern_ipc_perm *)*size);
L
Linus Torvalds 已提交
299 300 301 302 303 304 305 306 307 308 309
	for(i=size;i<newsize;i++) {
		new->p[i] = NULL;
	}
	old = ids->entries;

	/*
	 * Use rcu_assign_pointer() to make sure the memcpyed contents
	 * of the new array are visible before the new array becomes visible.
	 */
	rcu_assign_pointer(ids->entries, new);

P
Pavel Emelianov 已提交
310
	__ipc_fini_ids(ids, old);
L
Linus Torvalds 已提交
311 312 313 314 315 316 317 318 319 320 321 322 323 324
	return newsize;
}

/**
 *	ipc_addid 	-	add an IPC identifier
 *	@ids: IPC identifier set
 *	@new: new IPC permission set
 *	@size: new size limit for the id array
 *
 *	Add an entry 'new' to the IPC arrays. The permissions object is
 *	initialised and the first free entry is set up and the id assigned
 *	is returned. The list is returned in a locked state on success.
 *	On failure the list is not locked and -1 is returned.
 *
I
Ingo Molnar 已提交
325
 *	Called with ipc_ids.mutex held.
L
Linus Torvalds 已提交
326 327 328 329 330 331 332 333 334 335
 */
 
int ipc_addid(struct ipc_ids* ids, struct kern_ipc_perm* new, int size)
{
	int id;

	size = grow_ary(ids,size);

	/*
	 * rcu_dereference()() is not needed here since
I
Ingo Molnar 已提交
336
	 * ipc_ids.mutex is held
L
Linus Torvalds 已提交
337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371
	 */
	for (id = 0; id < size; id++) {
		if(ids->entries->p[id] == NULL)
			goto found;
	}
	return -1;
found:
	ids->in_use++;
	if (id > ids->max_id)
		ids->max_id = id;

	new->cuid = new->uid = current->euid;
	new->gid = new->cgid = current->egid;

	new->seq = ids->seq++;
	if(ids->seq > ids->seq_max)
		ids->seq = 0;

	spin_lock_init(&new->lock);
	new->deleted = 0;
	rcu_read_lock();
	spin_lock(&new->lock);
	ids->entries->p[id] = new;
	return id;
}

/**
 *	ipc_rmid	-	remove an IPC identifier
 *	@ids: identifier set
 *	@id: Identifier to remove
 *
 *	The identifier must be valid, and in use. The kernel will panic if
 *	fed an invalid identifier. The entry is removed and internal
 *	variables recomputed. The object associated with the identifier
 *	is returned.
I
Ingo Molnar 已提交
372
 *	ipc_ids.mutex and the spinlock for this ID is hold before this function
L
Linus Torvalds 已提交
373 374 375 376 377 378 379
 *	is called, and remain locked on the exit.
 */
 
struct kern_ipc_perm* ipc_rmid(struct ipc_ids* ids, int id)
{
	struct kern_ipc_perm* p;
	int lid = id % SEQ_MULTIPLIER;
380
	BUG_ON(lid >= ids->entries->size);
L
Linus Torvalds 已提交
381 382 383

	/* 
	 * do not need a rcu_dereference()() here to force ordering
I
Ingo Molnar 已提交
384
	 * on Alpha, since the ipc_ids.mutex is held.
L
Linus Torvalds 已提交
385 386 387
	 */	
	p = ids->entries->p[lid];
	ids->entries->p[lid] = NULL;
388
	BUG_ON(p==NULL);
L
Linus Torvalds 已提交
389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425
	ids->in_use--;

	if (lid == ids->max_id) {
		do {
			lid--;
			if(lid == -1)
				break;
		} while (ids->entries->p[lid] == NULL);
		ids->max_id = lid;
	}
	p->deleted = 1;
	return p;
}

/**
 *	ipc_alloc	-	allocate ipc space
 *	@size: size desired
 *
 *	Allocate memory from the appropriate pools and return a pointer to it.
 *	NULL is returned if the allocation fails
 */
 
void* ipc_alloc(int size)
{
	void* out;
	if(size > PAGE_SIZE)
		out = vmalloc(size);
	else
		out = kmalloc(size, GFP_KERNEL);
	return out;
}

/**
 *	ipc_free        -       free ipc space
 *	@ptr: pointer returned by ipc_alloc
 *	@size: size of block
 *
426
 *	Free a block created with ipc_alloc(). The caller must know the size
L
Linus Torvalds 已提交
427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522
 *	used in the allocation call.
 */

void ipc_free(void* ptr, int size)
{
	if(size > PAGE_SIZE)
		vfree(ptr);
	else
		kfree(ptr);
}

/*
 * rcu allocations:
 * There are three headers that are prepended to the actual allocation:
 * - during use: ipc_rcu_hdr.
 * - during the rcu grace period: ipc_rcu_grace.
 * - [only if vmalloc]: ipc_rcu_sched.
 * Their lifetime doesn't overlap, thus the headers share the same memory.
 * Unlike a normal union, they are right-aligned, thus some container_of
 * forward/backward casting is necessary:
 */
struct ipc_rcu_hdr
{
	int refcount;
	int is_vmalloc;
	void *data[0];
};


struct ipc_rcu_grace
{
	struct rcu_head rcu;
	/* "void *" makes sure alignment of following data is sane. */
	void *data[0];
};

struct ipc_rcu_sched
{
	struct work_struct work;
	/* "void *" makes sure alignment of following data is sane. */
	void *data[0];
};

#define HDRLEN_KMALLOC		(sizeof(struct ipc_rcu_grace) > sizeof(struct ipc_rcu_hdr) ? \
					sizeof(struct ipc_rcu_grace) : sizeof(struct ipc_rcu_hdr))
#define HDRLEN_VMALLOC		(sizeof(struct ipc_rcu_sched) > HDRLEN_KMALLOC ? \
					sizeof(struct ipc_rcu_sched) : HDRLEN_KMALLOC)

static inline int rcu_use_vmalloc(int size)
{
	/* Too big for a single page? */
	if (HDRLEN_KMALLOC + size > PAGE_SIZE)
		return 1;
	return 0;
}

/**
 *	ipc_rcu_alloc	-	allocate ipc and rcu space 
 *	@size: size desired
 *
 *	Allocate memory for the rcu header structure +  the object.
 *	Returns the pointer to the object.
 *	NULL is returned if the allocation fails. 
 */
 
void* ipc_rcu_alloc(int size)
{
	void* out;
	/* 
	 * We prepend the allocation with the rcu struct, and
	 * workqueue if necessary (for vmalloc). 
	 */
	if (rcu_use_vmalloc(size)) {
		out = vmalloc(HDRLEN_VMALLOC + size);
		if (out) {
			out += HDRLEN_VMALLOC;
			container_of(out, struct ipc_rcu_hdr, data)->is_vmalloc = 1;
			container_of(out, struct ipc_rcu_hdr, data)->refcount = 1;
		}
	} else {
		out = kmalloc(HDRLEN_KMALLOC + size, GFP_KERNEL);
		if (out) {
			out += HDRLEN_KMALLOC;
			container_of(out, struct ipc_rcu_hdr, data)->is_vmalloc = 0;
			container_of(out, struct ipc_rcu_hdr, data)->refcount = 1;
		}
	}

	return out;
}

void ipc_rcu_getref(void *ptr)
{
	container_of(ptr, struct ipc_rcu_hdr, data)->refcount++;
}

523 524 525 526 527
static void ipc_do_vfree(struct work_struct *work)
{
	vfree(container_of(work, struct ipc_rcu_sched, work));
}

L
Linus Torvalds 已提交
528
/**
529 530
 * ipc_schedule_free - free ipc + rcu space
 * @head: RCU callback structure for queued work
L
Linus Torvalds 已提交
531 532
 * 
 * Since RCU callback function is called in bh,
533
 * we need to defer the vfree to schedule_work().
L
Linus Torvalds 已提交
534 535 536 537 538 539 540 541
 */
static void ipc_schedule_free(struct rcu_head *head)
{
	struct ipc_rcu_grace *grace =
		container_of(head, struct ipc_rcu_grace, rcu);
	struct ipc_rcu_sched *sched =
			container_of(&(grace->data[0]), struct ipc_rcu_sched, data[0]);

542
	INIT_WORK(&sched->work, ipc_do_vfree);
L
Linus Torvalds 已提交
543 544 545 546
	schedule_work(&sched->work);
}

/**
547 548
 * ipc_immediate_free - free ipc + rcu space
 * @head: RCU callback structure that contains pointer to be freed
L
Linus Torvalds 已提交
549
 *
550
 * Free from the RCU callback context.
L
Linus Torvalds 已提交
551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583
 */
static void ipc_immediate_free(struct rcu_head *head)
{
	struct ipc_rcu_grace *free =
		container_of(head, struct ipc_rcu_grace, rcu);
	kfree(free);
}

void ipc_rcu_putref(void *ptr)
{
	if (--container_of(ptr, struct ipc_rcu_hdr, data)->refcount > 0)
		return;

	if (container_of(ptr, struct ipc_rcu_hdr, data)->is_vmalloc) {
		call_rcu(&container_of(ptr, struct ipc_rcu_grace, data)->rcu,
				ipc_schedule_free);
	} else {
		call_rcu(&container_of(ptr, struct ipc_rcu_grace, data)->rcu,
				ipc_immediate_free);
	}
}

/**
 *	ipcperms	-	check IPC permissions
 *	@ipcp: IPC permission set
 *	@flag: desired permission set.
 *
 *	Check user, group, other permissions for access
 *	to ipc resources. return 0 if allowed
 */
 
int ipcperms (struct kern_ipc_perm *ipcp, short flag)
{	/* flag will most probably be 0 or S_...UGO from <linux/stat.h> */
S
Steve Grubb 已提交
584
	int requested_mode, granted_mode, err;
L
Linus Torvalds 已提交
585

S
Steve Grubb 已提交
586 587
	if (unlikely((err = audit_ipc_obj(ipcp))))
		return err;
L
Linus Torvalds 已提交
588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611
	requested_mode = (flag >> 6) | (flag >> 3) | flag;
	granted_mode = ipcp->mode;
	if (current->euid == ipcp->cuid || current->euid == ipcp->uid)
		granted_mode >>= 6;
	else if (in_group_p(ipcp->cgid) || in_group_p(ipcp->gid))
		granted_mode >>= 3;
	/* is there some bit set in requested_mode but not in granted_mode? */
	if ((requested_mode & ~granted_mode & 0007) && 
	    !capable(CAP_IPC_OWNER))
		return -1;

	return security_ipc_permission(ipcp, flag);
}

/*
 * Functions to convert between the kern_ipc_perm structure and the
 * old/new ipc_perm structures
 */

/**
 *	kernel_to_ipc64_perm	-	convert kernel ipc permissions to user
 *	@in: kernel permissions
 *	@out: new style IPC permissions
 *
612 613
 *	Turn the kernel object @in into a set of permissions descriptions
 *	for returning to userspace (@out).
L
Linus Torvalds 已提交
614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632
 */
 

void kernel_to_ipc64_perm (struct kern_ipc_perm *in, struct ipc64_perm *out)
{
	out->key	= in->key;
	out->uid	= in->uid;
	out->gid	= in->gid;
	out->cuid	= in->cuid;
	out->cgid	= in->cgid;
	out->mode	= in->mode;
	out->seq	= in->seq;
}

/**
 *	ipc64_perm_to_ipc_perm	-	convert old ipc permissions to new
 *	@in: new style IPC permissions
 *	@out: old style IPC permissions
 *
633 634
 *	Turn the new style permissions object @in into a compatibility
 *	object and store it into the @out pointer.
L
Linus Torvalds 已提交
635 636 637 638 639 640 641 642 643 644 645 646 647 648 649
 */
 
void ipc64_perm_to_ipc_perm (struct ipc64_perm *in, struct ipc_perm *out)
{
	out->key	= in->key;
	SET_UID(out->uid, in->uid);
	SET_GID(out->gid, in->gid);
	SET_UID(out->cuid, in->cuid);
	SET_GID(out->cgid, in->cgid);
	out->mode	= in->mode;
	out->seq	= in->seq;
}

/*
 * So far only shm_get_stat() calls ipc_get() via shm_get(), so ipc_get()
I
Ingo Molnar 已提交
650 651
 * is called with shm_ids.mutex locked.  Since grow_ary() is also called with
 * shm_ids.mutex down(for Shared Memory), there is no need to add read
L
Linus Torvalds 已提交
652 653 654
 * barriers here to gurantee the writes in grow_ary() are seen in order 
 * here (for Alpha).
 *
I
Ingo Molnar 已提交
655 656
 * However ipc_get() itself does not necessary require ipc_ids.mutex down. So
 * if in the future ipc_get() is used by other places without ipc_ids.mutex
L
Linus Torvalds 已提交
657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730
 * down, then ipc_get() needs read memery barriers as ipc_lock() does.
 */
struct kern_ipc_perm* ipc_get(struct ipc_ids* ids, int id)
{
	struct kern_ipc_perm* out;
	int lid = id % SEQ_MULTIPLIER;
	if(lid >= ids->entries->size)
		return NULL;
	out = ids->entries->p[lid];
	return out;
}

struct kern_ipc_perm* ipc_lock(struct ipc_ids* ids, int id)
{
	struct kern_ipc_perm* out;
	int lid = id % SEQ_MULTIPLIER;
	struct ipc_id_ary* entries;

	rcu_read_lock();
	entries = rcu_dereference(ids->entries);
	if(lid >= entries->size) {
		rcu_read_unlock();
		return NULL;
	}
	out = entries->p[lid];
	if(out == NULL) {
		rcu_read_unlock();
		return NULL;
	}
	spin_lock(&out->lock);
	
	/* ipc_rmid() may have already freed the ID while ipc_lock
	 * was spinning: here verify that the structure is still valid
	 */
	if (out->deleted) {
		spin_unlock(&out->lock);
		rcu_read_unlock();
		return NULL;
	}
	return out;
}

void ipc_lock_by_ptr(struct kern_ipc_perm *perm)
{
	rcu_read_lock();
	spin_lock(&perm->lock);
}

void ipc_unlock(struct kern_ipc_perm* perm)
{
	spin_unlock(&perm->lock);
	rcu_read_unlock();
}

int ipc_buildid(struct ipc_ids* ids, int id, int seq)
{
	return SEQ_MULTIPLIER*seq + id;
}

int ipc_checkid(struct ipc_ids* ids, struct kern_ipc_perm* ipcp, int uid)
{
	if(uid/SEQ_MULTIPLIER != ipcp->seq)
		return 1;
	return 0;
}

#ifdef __ARCH_WANT_IPC_PARSE_VERSION


/**
 *	ipc_parse_version	-	IPC call version
 *	@cmd: pointer to command
 *
 *	Return IPC_64 for new style IPC and IPC_OLD for old style IPC. 
731
 *	The @cmd value is turned from an encoding command and version into
L
Linus Torvalds 已提交
732 733 734 735 736 737 738 739 740 741 742 743 744 745
 *	just the command code.
 */
 
int ipc_parse_version (int *cmd)
{
	if (*cmd & IPC_64) {
		*cmd ^= IPC_64;
		return IPC_64;
	} else {
		return IPC_OLD;
	}
}

#endif /* __ARCH_WANT_IPC_PARSE_VERSION */
746 747

#ifdef CONFIG_PROC_FS
748 749 750 751 752
struct ipc_proc_iter {
	struct ipc_namespace *ns;
	struct ipc_proc_iface *iface;
};

753 754
static void *sysvipc_proc_next(struct seq_file *s, void *it, loff_t *pos)
{
755 756
	struct ipc_proc_iter *iter = s->private;
	struct ipc_proc_iface *iface = iter->iface;
757 758
	struct kern_ipc_perm *ipc = it;
	loff_t p;
K
Kirill Korotaev 已提交
759 760
	struct ipc_ids *ids;

761
	ids = iter->ns->ids[iface->ids];
762 763 764 765 766 767 768 769 770

	/* If we had an ipc id locked before, unlock it */
	if (ipc && ipc != SEQ_START_TOKEN)
		ipc_unlock(ipc);

	/*
	 * p = *pos - 1 (because id 0 starts at position 1)
	 *          + 1 (because we increment the position by one)
	 */
K
Kirill Korotaev 已提交
771 772
	for (p = *pos; p <= ids->max_id; p++) {
		if ((ipc = ipc_lock(ids, p)) != NULL) {
773 774 775 776 777 778 779 780 781 782 783 784 785 786 787
			*pos = p + 1;
			return ipc;
		}
	}

	/* Out of range - return NULL to terminate iteration */
	return NULL;
}

/*
 * File positions: pos 0 -> header, pos n -> ipc id + 1.
 * SeqFile iterator: iterator value locked shp or SEQ_TOKEN_START.
 */
static void *sysvipc_proc_start(struct seq_file *s, loff_t *pos)
{
788 789
	struct ipc_proc_iter *iter = s->private;
	struct ipc_proc_iface *iface = iter->iface;
790 791
	struct kern_ipc_perm *ipc;
	loff_t p;
K
Kirill Korotaev 已提交
792 793
	struct ipc_ids *ids;

794
	ids = iter->ns->ids[iface->ids];
795 796 797 798 799

	/*
	 * Take the lock - this will be released by the corresponding
	 * call to stop().
	 */
K
Kirill Korotaev 已提交
800
	mutex_lock(&ids->mutex);
801 802 803 804 805 806 807 808 809 810

	/* pos < 0 is invalid */
	if (*pos < 0)
		return NULL;

	/* pos == 0 means header */
	if (*pos == 0)
		return SEQ_START_TOKEN;

	/* Find the (pos-1)th ipc */
K
Kirill Korotaev 已提交
811 812
	for (p = *pos - 1; p <= ids->max_id; p++) {
		if ((ipc = ipc_lock(ids, p)) != NULL) {
813 814 815 816 817 818 819 820 821 822
			*pos = p + 1;
			return ipc;
		}
	}
	return NULL;
}

static void sysvipc_proc_stop(struct seq_file *s, void *it)
{
	struct kern_ipc_perm *ipc = it;
823 824
	struct ipc_proc_iter *iter = s->private;
	struct ipc_proc_iface *iface = iter->iface;
K
Kirill Korotaev 已提交
825
	struct ipc_ids *ids;
826 827 828 829 830

	/* If we had a locked segment, release it */
	if (ipc && ipc != SEQ_START_TOKEN)
		ipc_unlock(ipc);

831
	ids = iter->ns->ids[iface->ids];
832
	/* Release the lock we took in start() */
K
Kirill Korotaev 已提交
833
	mutex_unlock(&ids->mutex);
834 835 836 837
}

static int sysvipc_proc_show(struct seq_file *s, void *it)
{
838 839
	struct ipc_proc_iter *iter = s->private;
	struct ipc_proc_iface *iface = iter->iface;
840 841 842 843 844 845 846 847 848 849 850 851 852 853

	if (it == SEQ_START_TOKEN)
		return seq_puts(s, iface->header);

	return iface->show(s, it);
}

static struct seq_operations sysvipc_proc_seqops = {
	.start = sysvipc_proc_start,
	.stop  = sysvipc_proc_stop,
	.next  = sysvipc_proc_next,
	.show  = sysvipc_proc_show,
};

854 855
static int sysvipc_proc_open(struct inode *inode, struct file *file)
{
856 857
	int ret;
	struct seq_file *seq;
858 859 860 861 862 863
	struct ipc_proc_iter *iter;

	ret = -ENOMEM;
	iter = kmalloc(sizeof(*iter), GFP_KERNEL);
	if (!iter)
		goto out;
864 865

	ret = seq_open(file, &sysvipc_proc_seqops);
866 867 868 869 870 871 872 873 874
	if (ret)
		goto out_kfree;

	seq = file->private_data;
	seq->private = iter;

	iter->iface = PDE(inode)->data;
	iter->ns    = get_ipc_ns(current->nsproxy->ipc_ns);
out:
875
	return ret;
876 877 878 879 880 881 882 883 884 885 886
out_kfree:
	kfree(iter);
	goto out;
}

static int sysvipc_proc_release(struct inode *inode, struct file *file)
{
	struct seq_file *seq = file->private_data;
	struct ipc_proc_iter *iter = seq->private;
	put_ipc_ns(iter->ns);
	return seq_release_private(inode, file);
887 888
}

889
static const struct file_operations sysvipc_proc_fops = {
890 891 892
	.open    = sysvipc_proc_open,
	.read    = seq_read,
	.llseek  = seq_lseek,
893
	.release = sysvipc_proc_release,
894 895
};
#endif /* CONFIG_PROC_FS */