taskstats.c 12.5 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
/*
 * taskstats.c - Export per-task statistics to userland
 *
 * Copyright (C) Shailabh Nagar, IBM Corp. 2006
 *           (C) Balbir Singh,   IBM Corp. 2006
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 */

#include <linux/kernel.h>
#include <linux/taskstats_kern.h>
21
#include <linux/tsacct_kern.h>
22
#include <linux/delayacct.h>
23
#include <linux/tsacct_kern.h>
24 25
#include <linux/cpumask.h>
#include <linux/percpu.h>
26 27 28
#include <net/genetlink.h>
#include <asm/atomic.h>

29 30 31 32 33 34
/*
 * Maximum length of a cpumask that can be specified in
 * the TASKSTATS_CMD_ATTR_REGISTER/DEREGISTER_CPUMASK attribute
 */
#define TASKSTATS_CPUMASK_MAXLEN	(100+6*NR_CPUS)

35 36 37 38 39 40 41 42 43 44 45 46 47 48 49
static DEFINE_PER_CPU(__u32, taskstats_seqnum) = { 0 };
static int family_registered;
kmem_cache_t *taskstats_cache;

static struct genl_family family = {
	.id		= GENL_ID_GENERATE,
	.name		= TASKSTATS_GENL_NAME,
	.version	= TASKSTATS_GENL_VERSION,
	.maxattr	= TASKSTATS_CMD_ATTR_MAX,
};

static struct nla_policy taskstats_cmd_get_policy[TASKSTATS_CMD_ATTR_MAX+1]
__read_mostly = {
	[TASKSTATS_CMD_ATTR_PID]  = { .type = NLA_U32 },
	[TASKSTATS_CMD_ATTR_TGID] = { .type = NLA_U32 },
50 51 52 53 54 55
	[TASKSTATS_CMD_ATTR_REGISTER_CPUMASK] = { .type = NLA_STRING },
	[TASKSTATS_CMD_ATTR_DEREGISTER_CPUMASK] = { .type = NLA_STRING },};

struct listener {
	struct list_head list;
	pid_t pid;
56
	char valid;
57 58
};

59 60 61 62 63 64 65 66 67 68 69
struct listener_list {
	struct rw_semaphore sem;
	struct list_head list;
};
static DEFINE_PER_CPU(struct listener_list, listener_array);

enum actions {
	REGISTER,
	DEREGISTER,
	CPU_DONT_CARE
};
70 71 72 73 74 75 76 77 78 79

static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp,
			void **replyp, size_t size)
{
	struct sk_buff *skb;
	void *reply;

	/*
	 * If new attributes are added, please revisit this allocation
	 */
80
	skb = genlmsg_new(size, GFP_KERNEL);
81 82 83 84 85 86 87
	if (!skb)
		return -ENOMEM;

	if (!info) {
		int seq = get_cpu_var(taskstats_seqnum)++;
		put_cpu_var(taskstats_seqnum);

88
		reply = genlmsg_put(skb, 0, seq, &family, 0, cmd);
89
	} else
90
		reply = genlmsg_put_reply(skb, info, &family, 0, cmd);
91 92 93 94 95 96 97 98 99 100
	if (reply == NULL) {
		nlmsg_free(skb);
		return -EINVAL;
	}

	*skbp = skb;
	*replyp = reply;
	return 0;
}

101 102 103 104
/*
 * Send taskstats data in @skb to listener with nl_pid @pid
 */
static int send_reply(struct sk_buff *skb, pid_t pid)
105 106
{
	struct genlmsghdr *genlhdr = nlmsg_data((struct nlmsghdr *)skb->data);
107
	void *reply = genlmsg_data(genlhdr);
108 109 110 111 112 113 114 115 116 117 118
	int rc;

	rc = genlmsg_end(skb, reply);
	if (rc < 0) {
		nlmsg_free(skb);
		return rc;
	}

	return genlmsg_unicast(skb, pid);
}

119 120 121
/*
 * Send taskstats data in @skb to listeners registered for @cpu's exit data
 */
122
static void send_cpu_listeners(struct sk_buff *skb, unsigned int cpu)
123 124 125 126 127 128
{
	struct genlmsghdr *genlhdr = nlmsg_data((struct nlmsghdr *)skb->data);
	struct listener_list *listeners;
	struct listener *s, *tmp;
	struct sk_buff *skb_next, *skb_cur = skb;
	void *reply = genlmsg_data(genlhdr);
129
	int rc, delcount = 0;
130 131 132 133

	rc = genlmsg_end(skb, reply);
	if (rc < 0) {
		nlmsg_free(skb);
134
		return;
135 136 137 138
	}

	rc = 0;
	listeners = &per_cpu(listener_array, cpu);
139
	down_read(&listeners->sem);
140
	list_for_each_entry(s, &listeners->list, list) {
141 142 143
		skb_next = NULL;
		if (!list_is_last(&s->list, &listeners->list)) {
			skb_next = skb_clone(skb_cur, GFP_KERNEL);
144
			if (!skb_next)
145 146
				break;
		}
147 148
		rc = genlmsg_unicast(skb_cur, s->pid);
		if (rc == -ECONNREFUSED) {
149 150
			s->valid = 0;
			delcount++;
151 152 153
		}
		skb_cur = skb_next;
	}
154
	up_read(&listeners->sem);
155

156 157 158
	if (skb_cur)
		nlmsg_free(skb_cur);

159
	if (!delcount)
160
		return;
161 162 163 164 165 166 167 168 169 170

	/* Delete invalidated entries */
	down_write(&listeners->sem);
	list_for_each_entry_safe(s, tmp, &listeners->list, list) {
		if (!s->valid) {
			list_del(&s->list);
			kfree(s);
		}
	}
	up_write(&listeners->sem);
171 172
}

173
static int fill_pid(pid_t pid, struct task_struct *tsk,
174 175
		struct taskstats *stats)
{
176
	int rc = 0;
177

178 179
	if (!tsk) {
		rcu_read_lock();
180
		tsk = find_task_by_pid(pid);
181 182 183 184
		if (tsk)
			get_task_struct(tsk);
		rcu_read_unlock();
		if (!tsk)
185 186 187 188 189 190 191 192
			return -ESRCH;
	} else
		get_task_struct(tsk);

	/*
	 * Each accounting subsystem adds calls to its functions to
	 * fill in relevant parts of struct taskstsats as follows
	 *
193
	 *	per-task-foo(stats, tsk);
194 195
	 */

196
	delayacct_add_tsk(stats, tsk);
197 198

	/* fill in basic acct fields */
199
	stats->version = TASKSTATS_VERSION;
200
	bacct_add_tsk(stats, tsk);
201

202 203 204
	/* fill in extended acct fields */
	xacct_add_tsk(stats, tsk);

205
	/* Define err: label here if needed */
206 207 208 209 210
	put_task_struct(tsk);
	return rc;

}

211
static int fill_tgid(pid_t tgid, struct task_struct *first,
212 213
		struct taskstats *stats)
{
214
	struct task_struct *tsk;
215
	unsigned long flags;
216
	int rc = -ESRCH;
217

218 219 220 221
	/*
	 * Add additional stats from live tasks except zombie thread group
	 * leaders who are already counted with the dead tasks
	 */
222 223
	rcu_read_lock();
	if (!first)
224
		first = find_task_by_pid(tgid);
225

226 227
	if (!first || !lock_task_sighand(first, &flags))
		goto out;
228

229 230
	if (first->signal->stats)
		memcpy(stats, first->signal->stats, sizeof(*stats));
231

232
	tsk = first;
233
	do {
234
		if (tsk->exit_state)
235
			continue;
236
		/*
237
		 * Accounting subsystem can call its functions here to
238 239
		 * fill in relevant parts of struct taskstsats as follows
		 *
240
		 *	per-task-foo(stats, tsk);
241
		 */
242
		delayacct_add_tsk(stats, tsk);
243

244
	} while_each_thread(first, tsk);
245

246 247 248 249 250 251
	unlock_task_sighand(first, &flags);
	rc = 0;
out:
	rcu_read_unlock();

	stats->version = TASKSTATS_VERSION;
252
	/*
253 254
	 * Accounting subsytems can also add calls here to modify
	 * fields of taskstats.
255
	 */
256
	return rc;
257 258 259 260 261 262 263
}


static void fill_tgid_exit(struct task_struct *tsk)
{
	unsigned long flags;

264
	spin_lock_irqsave(&tsk->sighand->siglock, flags);
265 266 267 268 269 270 271 272 273 274 275
	if (!tsk->signal->stats)
		goto ret;

	/*
	 * Each accounting subsystem calls its functions here to
	 * accumalate its per-task stats for tsk, into the per-tgid structure
	 *
	 *	per-task-foo(tsk->signal->stats, tsk);
	 */
	delayacct_add_tsk(tsk->signal->stats, tsk);
ret:
276
	spin_unlock_irqrestore(&tsk->sighand->siglock, flags);
277
	return;
278 279
}

280 281 282 283 284 285
static int add_del_listener(pid_t pid, cpumask_t *maskp, int isadd)
{
	struct listener_list *listeners;
	struct listener *s, *tmp;
	unsigned int cpu;
	cpumask_t mask = *maskp;
286

287 288 289 290 291 292 293 294 295 296 297
	if (!cpus_subset(mask, cpu_possible_map))
		return -EINVAL;

	if (isadd == REGISTER) {
		for_each_cpu_mask(cpu, mask) {
			s = kmalloc_node(sizeof(struct listener), GFP_KERNEL,
					 cpu_to_node(cpu));
			if (!s)
				goto cleanup;
			s->pid = pid;
			INIT_LIST_HEAD(&s->list);
298
			s->valid = 1;
299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347

			listeners = &per_cpu(listener_array, cpu);
			down_write(&listeners->sem);
			list_add(&s->list, &listeners->list);
			up_write(&listeners->sem);
		}
		return 0;
	}

	/* Deregister or cleanup */
cleanup:
	for_each_cpu_mask(cpu, mask) {
		listeners = &per_cpu(listener_array, cpu);
		down_write(&listeners->sem);
		list_for_each_entry_safe(s, tmp, &listeners->list, list) {
			if (s->pid == pid) {
				list_del(&s->list);
				kfree(s);
				break;
			}
		}
		up_write(&listeners->sem);
	}
	return 0;
}

static int parse(struct nlattr *na, cpumask_t *mask)
{
	char *data;
	int len;
	int ret;

	if (na == NULL)
		return 1;
	len = nla_len(na);
	if (len > TASKSTATS_CPUMASK_MAXLEN)
		return -E2BIG;
	if (len < 1)
		return -EINVAL;
	data = kmalloc(len, GFP_KERNEL);
	if (!data)
		return -ENOMEM;
	nla_strlcpy(data, na, len);
	ret = cpulist_parse(data, *mask);
	kfree(data);
	return ret;
}

static int taskstats_user_cmd(struct sk_buff *skb, struct genl_info *info)
348 349 350 351 352 353 354
{
	int rc = 0;
	struct sk_buff *rep_skb;
	struct taskstats stats;
	void *reply;
	size_t size;
	struct nlattr *na;
355 356 357 358 359 360 361 362 363 364 365 366 367
	cpumask_t mask;

	rc = parse(info->attrs[TASKSTATS_CMD_ATTR_REGISTER_CPUMASK], &mask);
	if (rc < 0)
		return rc;
	if (rc == 0)
		return add_del_listener(info->snd_pid, &mask, REGISTER);

	rc = parse(info->attrs[TASKSTATS_CMD_ATTR_DEREGISTER_CPUMASK], &mask);
	if (rc < 0)
		return rc;
	if (rc == 0)
		return add_del_listener(info->snd_pid, &mask, DEREGISTER);
368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406

	/*
	 * Size includes space for nested attributes
	 */
	size = nla_total_size(sizeof(u32)) +
		nla_total_size(sizeof(struct taskstats)) + nla_total_size(0);

	memset(&stats, 0, sizeof(stats));
	rc = prepare_reply(info, TASKSTATS_CMD_NEW, &rep_skb, &reply, size);
	if (rc < 0)
		return rc;

	if (info->attrs[TASKSTATS_CMD_ATTR_PID]) {
		u32 pid = nla_get_u32(info->attrs[TASKSTATS_CMD_ATTR_PID]);
		rc = fill_pid(pid, NULL, &stats);
		if (rc < 0)
			goto err;

		na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_PID);
		NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_PID, pid);
		NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS,
				stats);
	} else if (info->attrs[TASKSTATS_CMD_ATTR_TGID]) {
		u32 tgid = nla_get_u32(info->attrs[TASKSTATS_CMD_ATTR_TGID]);
		rc = fill_tgid(tgid, NULL, &stats);
		if (rc < 0)
			goto err;

		na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_TGID);
		NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_TGID, tgid);
		NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS,
				stats);
	} else {
		rc = -EINVAL;
		goto err;
	}

	nla_nest_end(rep_skb, na);

407
	return send_reply(rep_skb, info->snd_pid);
408 409

nla_put_failure:
410
	rc = genlmsg_cancel(rep_skb, reply);
411 412 413 414 415
err:
	nlmsg_free(rep_skb);
	return rc;
}

416 417 418 419 420 421 422 423 424 425 426 427
void taskstats_exit_alloc(struct taskstats **ptidstats, unsigned int *mycpu)
{
	struct listener_list *listeners;
	struct taskstats *tmp;
	/*
	 * This is the cpu on which the task is exiting currently and will
	 * be the one for which the exit event is sent, even if the cpu
	 * on which this function is running changes later.
	 */
	*mycpu = raw_smp_processor_id();

	*ptidstats = NULL;
428
	tmp = kmem_cache_zalloc(taskstats_cache, GFP_KERNEL);
429 430 431 432 433 434 435 436 437 438 439 440 441
	if (!tmp)
		return;

	listeners = &per_cpu(listener_array, *mycpu);
	down_read(&listeners->sem);
	if (!list_empty(&listeners->list)) {
		*ptidstats = tmp;
		tmp = NULL;
	}
	up_read(&listeners->sem);
	kfree(tmp);
}

442 443
/* Send pid data out on exit */
void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats,
444
			int group_dead, unsigned int mycpu)
445 446 447 448 449 450 451 452
{
	int rc;
	struct sk_buff *rep_skb;
	void *reply;
	size_t size;
	int is_thread_group;
	struct nlattr *na;

453
	if (!family_registered)
454 455 456 457 458 459 460 461
		return;

	/*
	 * Size includes space for nested attributes
	 */
	size = nla_total_size(sizeof(u32)) +
		nla_total_size(sizeof(struct taskstats)) + nla_total_size(0);

462
	is_thread_group = (tsk->signal->stats != NULL);
463 464 465 466 467 468 469 470 471
	if (is_thread_group) {
		/* PID + STATS + TGID + STATS */
		size = 2 * size;
		/* fill the tsk->signal->stats structure */
		fill_tgid_exit(tsk);
	}

	if (!tidstats)
		return;
472 473 474 475 476 477 478 479 480 481 482 483 484 485 486

	rc = prepare_reply(NULL, TASKSTATS_CMD_NEW, &rep_skb, &reply, size);
	if (rc < 0)
		goto ret;

	rc = fill_pid(tsk->pid, tsk, tidstats);
	if (rc < 0)
		goto err_skb;

	na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_PID);
	NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_PID, (u32)tsk->pid);
	NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS,
			*tidstats);
	nla_nest_end(rep_skb, na);

487 488
	if (!is_thread_group)
		goto send;
489 490

	/*
491
	 * Doesn't matter if tsk is the leader or the last group member leaving
492
	 */
493 494
	if (!group_dead)
		goto send;
495 496 497

	na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_TGID);
	NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_TGID, (u32)tsk->tgid);
498
	/* No locking needed for tsk->signal->stats since group is dead */
499
	NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS,
500
			*tsk->signal->stats);
501 502
	nla_nest_end(rep_skb, na);

503
send:
504
	send_cpu_listeners(rep_skb, mycpu);
505
	return;
506 507 508 509 510 511 512 513 514 515 516

nla_put_failure:
	genlmsg_cancel(rep_skb, reply);
err_skb:
	nlmsg_free(rep_skb);
ret:
	return;
}

static struct genl_ops taskstats_ops = {
	.cmd		= TASKSTATS_CMD_GET,
517
	.doit		= taskstats_user_cmd,
518 519 520 521 522 523
	.policy		= taskstats_cmd_get_policy,
};

/* Needed early in initialization */
void __init taskstats_init_early(void)
{
524 525
	unsigned int i;

526 527 528
	taskstats_cache = kmem_cache_create("taskstats_cache",
						sizeof(struct taskstats),
						0, SLAB_PANIC, NULL, NULL);
529 530 531 532
	for_each_possible_cpu(i) {
		INIT_LIST_HEAD(&(per_cpu(listener_array, i).list));
		init_rwsem(&(per_cpu(listener_array, i).sem));
	}
533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558
}

static int __init taskstats_init(void)
{
	int rc;

	rc = genl_register_family(&family);
	if (rc)
		return rc;

	rc = genl_register_ops(&family, &taskstats_ops);
	if (rc < 0)
		goto err;

	family_registered = 1;
	return 0;
err:
	genl_unregister_family(&family);
	return rc;
}

/*
 * late initcall ensures initialization of statistics collection
 * mechanisms precedes initialization of the taskstats interface
 */
late_initcall(taskstats_init);