[linux-2.6-block.git] / drivers / staging / android / lowmemorykiller.c

/* drivers/misc/lowmemorykiller.c
 *
 * The lowmemorykiller driver lets user-space specify a set of memory thresholds
 * where processes with a range of oom_score_adj values will get killed. Specify
 * the minimum oom_score_adj values in
 * /sys/module/lowmemorykiller/parameters/adj and the number of free pages in
 * /sys/module/lowmemorykiller/parameters/minfree. Both files take a comma
 * separated list of numbers in ascending order.
 *
 * For example, write "0,8" to /sys/module/lowmemorykiller/parameters/adj and
 * "1024,4096" to /sys/module/lowmemorykiller/parameters/minfree to kill
 * processes with a oom_score_adj value of 8 or higher when the free memory
 * drops below 4096 pages and kill processes with a oom_score_adj value of 0 or
 * higher when the free memory drops below 1024 pages.
 *
 * The driver considers memory used for caches to be free, but if a large
 * percentage of the cached memory is locked this can be very inaccurate
 * and processes may not get killed until the normal oom killer is triggered.
 *
 * Copyright (C) 2007-2008 Google, Inc.
 *
 * This software is licensed under the terms of the GNU General Public
 * License version 2, as published by the Free Software Foundation, and
 * may be copied, distributed, and modified under those terms.
 *
 * 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.
 *
 */

#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/oom.h>
#include <linux/sched.h>
#include <linux/swap.h>
#include <linux/rcupdate.h>
#include <linux/profile.h>
#include <linux/notifier.h>

static uint32_t lowmem_debug_level = 1;
static short lowmem_adj[6] = {
	0,
	1,
	6,
	12,
};
static int lowmem_adj_size = 4;
static int lowmem_minfree[6] = {
	3 * 512,	/* 6MB */
	2 * 1024,	/* 8MB */
	4 * 1024,	/* 16MB */
	16 * 1024,	/* 64MB */
};
static int lowmem_minfree_size = 4;

static unsigned long lowmem_deathpending_timeout;

#define lowmem_print(level, x...)			\
	do {						\
		if (lowmem_debug_level >= (level))	\
			pr_info(x);			\
	} while (0)

static unsigned long lowmem_count(struct shrinker *s,
				  struct shrink_control *sc)
{
	return global_page_state(NR_ACTIVE_ANON) +
		global_page_state(NR_ACTIVE_FILE) +
		global_page_state(NR_INACTIVE_ANON) +
		global_page_state(NR_INACTIVE_FILE);
}

static unsigned long lowmem_scan(struct shrinker *s, struct shrink_control *sc)
{
	struct task_struct *tsk;
	struct task_struct *selected = NULL;
	unsigned long rem = 0;
	int tasksize;
	int i;
	short min_score_adj = OOM_SCORE_ADJ_MAX + 1;
	int selected_tasksize = 0;
	short selected_oom_score_adj;
	int array_size = ARRAY_SIZE(lowmem_adj);
	int other_free = global_page_state(NR_FREE_PAGES) - totalreserve_pages;
	int other_file = global_page_state(NR_FILE_PAGES) -
						global_page_state(NR_SHMEM) -
						total_swapcache_pages();

	if (lowmem_adj_size < array_size)
		array_size = lowmem_adj_size;
	if (lowmem_minfree_size < array_size)
		array_size = lowmem_minfree_size;
	for (i = 0; i < array_size; i++) {
		if (other_free < lowmem_minfree[i] &&
		    other_file < lowmem_minfree[i]) {
			min_score_adj = lowmem_adj[i];
			break;
		}
	}

	lowmem_print(3, "lowmem_scan %lu, %x, ofree %d %d, ma %hd\n",
			sc->nr_to_scan, sc->gfp_mask, other_free,
			other_file, min_score_adj);

	if (min_score_adj == OOM_SCORE_ADJ_MAX + 1) {
		lowmem_print(5, "lowmem_scan %lu, %x, return 0\n",
			     sc->nr_to_scan, sc->gfp_mask);
		return 0;
	}

	selected_oom_score_adj = min_score_adj;

	rcu_read_lock();
	for_each_process(tsk) {
		struct task_struct *p;
		short oom_score_adj;

		if (tsk->flags & PF_KTHREAD)
			continue;

		p = find_lock_task_mm(tsk);
		if (!p)
			continue;

		if (test_tsk_thread_flag(p, TIF_MEMDIE) &&
		    time_before_eq(jiffies, lowmem_deathpending_timeout)) {
			task_unlock(p);
			rcu_read_unlock();
			return 0;
		}
		oom_score_adj = p->signal->oom_score_adj;
		if (oom_score_adj < min_score_adj) {
			task_unlock(p);
			continue;
		}
		tasksize = get_mm_rss(p->mm);
		task_unlock(p);
		if (tasksize <= 0)
			continue;
		if (selected) {
			if (oom_score_adj < selected_oom_score_adj)
				continue;
			if (oom_score_adj == selected_oom_score_adj &&
			    tasksize <= selected_tasksize)
				continue;
		}
		selected = p;
		selected_tasksize = tasksize;
		selected_oom_score_adj = oom_score_adj;
		lowmem_print(2, "select %d (%s), adj %hd, size %d, to kill\n",
			     p->pid, p->comm, oom_score_adj, tasksize);
	}
	if (selected) {
		lowmem_print(1, "send sigkill to %d (%s), adj %hd, size %d\n",
			     selected->pid, selected->comm,
			     selected_oom_score_adj, selected_tasksize);
		lowmem_deathpending_timeout = jiffies + HZ;
		/*
		 * FIXME: lowmemorykiller shouldn't abuse global OOM killer
		 * infrastructure. There is no real reason why the selected
		 * task should have access to the memory reserves.
		 */
		mark_tsk_oom_victim(selected);
		send_sig(SIGKILL, selected, 0);
		rem += selected_tasksize;
	}

	lowmem_print(4, "lowmem_scan %lu, %x, return %lu\n",
		     sc->nr_to_scan, sc->gfp_mask, rem);
	rcu_read_unlock();
	return rem;
}

static struct shrinker lowmem_shrinker = {
	.scan_objects = lowmem_scan,
	.count_objects = lowmem_count,
	.seeks = DEFAULT_SEEKS * 16
};

static int __init lowmem_init(void)
{
	register_shrinker(&lowmem_shrinker);
	return 0;
}

static void __exit lowmem_exit(void)
{
	unregister_shrinker(&lowmem_shrinker);
}

module_param_named(cost, lowmem_shrinker.seeks, int, S_IRUGO | S_IWUSR);
module_param_array_named(adj, lowmem_adj, short, &lowmem_adj_size,
			 S_IRUGO | S_IWUSR);
module_param_array_named(minfree, lowmem_minfree, uint, &lowmem_minfree_size,
			 S_IRUGO | S_IWUSR);
module_param_named(debug_level, lowmem_debug_level, uint, S_IRUGO | S_IWUSR);

module_init(lowmem_init);
module_exit(lowmem_exit);

MODULE_LICENSE("GPL");
Commit	Line	Data
355b0502 GKH	1	/* drivers/misc/lowmemorykiller.c
	2	*
	3	* The lowmemorykiller driver lets user-space specify a set of memory thresholds
940f77b0 DR	4	* where processes with a range of oom_score_adj values will get killed. Specify
	5	* the minimum oom_score_adj values in
	6	* /sys/module/lowmemorykiller/parameters/adj and the number of free pages in
	7	* /sys/module/lowmemorykiller/parameters/minfree. Both files take a comma
	8	* separated list of numbers in ascending order.
355b0502 GKH	9	*
355b0502 GKH	10	* For example, write "0,8" to /sys/module/lowmemorykiller/parameters/adj and
3bf5d65f	11	* "1024,4096" to /sys/module/lowmemorykiller/parameters/minfree to kill
940f77b0 DR	12	* processes with a oom_score_adj value of 8 or higher when the free memory
	13	* drops below 4096 pages and kill processes with a oom_score_adj value of 0 or
	14	* higher when the free memory drops below 1024 pages.
355b0502 GKH	15	*
	16	* The driver considers memory used for caches to be free, but if a large
	17	* percentage of the cached memory is locked this can be very inaccurate
	18	* and processes may not get killed until the normal oom killer is triggered.
	19	*
	20	* Copyright (C) 2007-2008 Google, Inc.
	21	*
	22	* This software is licensed under the terms of the GNU General Public
	23	* License version 2, as published by the Free Software Foundation, and
	24	* may be copied, distributed, and modified under those terms.
	25	*
	26	* This program is distributed in the hope that it will be useful,
	27	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	28	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	29	* GNU General Public License for more details.
	30	*
	31	*/
	32
6b83f915 DV	33	#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
6b83f915 DV	34
355b0502 GKH	35	#include <linux/module.h>
	36	#include <linux/kernel.h>
	37	#include <linux/mm.h>
	38	#include <linux/oom.h>
	39	#include <linux/sched.h>
eeb0f4f3	40	#include <linux/swap.h>
294b2711	41	#include <linux/rcupdate.h>
4755b72e SM	42	#include <linux/profile.h>
4755b72e SM	43	#include <linux/notifier.h>
355b0502	44
99150f6a	45	static uint32_t lowmem_debug_level = 1;
a9c58b90	46	static short lowmem_adj[6] = {
355b0502 GKH	47	0,
	48	1,
	49	6,
	50	12,
	51	};
	52	static int lowmem_adj_size = 4;
624b2250	53	static int lowmem_minfree[6] = {
355b0502 GKH	54	3 * 512, /* 6MB */
	55	2 * 1024, /* 8MB */
	56	4 * 1024, /* 16MB */
	57	16 * 1024, /* 64MB */
	58	};
	59	static int lowmem_minfree_size = 4;
	60
e5d7965f	61	static unsigned long lowmem_deathpending_timeout;
4755b72e	62
355b0502 GKH	63	#define lowmem_print(level, x...) \
	64	do { \
	65	if (lowmem_debug_level >= (level)) \
6b83f915	66	pr_info(x); \
355b0502 GKH	67	} while (0)
355b0502 GKH	68
7dc19d5a DC	69	static unsigned long lowmem_count(struct shrinker *s,
	70	struct shrink_control *sc)
	71	{
	72	return global_page_state(NR_ACTIVE_ANON) +
	73	global_page_state(NR_ACTIVE_FILE) +
	74	global_page_state(NR_INACTIVE_ANON) +
	75	global_page_state(NR_INACTIVE_FILE);
	76	}
	77
	78	static unsigned long lowmem_scan(struct shrinker s, struct shrink_control sc)
355b0502	79	{
95670001	80	struct task_struct *tsk;
355b0502	81	struct task_struct *selected = NULL;
7dc19d5a	82	unsigned long rem = 0;
355b0502 GKH	83	int tasksize;
355b0502 GKH	84	int i;
a9c58b90	85	short min_score_adj = OOM_SCORE_ADJ_MAX + 1;
355b0502	86	int selected_tasksize = 0;
a9c58b90	87	short selected_oom_score_adj;
355b0502	88	int array_size = ARRAY_SIZE(lowmem_adj);
eeb0f4f3	89	int other_free = global_page_state(NR_FREE_PAGES) - totalreserve_pages;
71b2c82b	90	int other_file = global_page_state(NR_FILE_PAGES) -
058dbde9 VM	91	global_page_state(NR_SHMEM) -
058dbde9 VM	92	total_swapcache_pages();
355b0502 GKH	93
	94	if (lowmem_adj_size < array_size)
	95	array_size = lowmem_adj_size;
	96	if (lowmem_minfree_size < array_size)
	97	array_size = lowmem_minfree_size;
	98	for (i = 0; i < array_size; i++) {
	99	if (other_free < lowmem_minfree[i] &&
	100	other_file < lowmem_minfree[i]) {
940f77b0	101	min_score_adj = lowmem_adj[i];
355b0502 GKH	102	break;
	103	}
	104	}
7dc19d5a DC	105
	106	lowmem_print(3, "lowmem_scan %lu, %x, ofree %d %d, ma %hd\n",
	107	sc->nr_to_scan, sc->gfp_mask, other_free,
	108	other_file, min_score_adj);
	109
	110	if (min_score_adj == OOM_SCORE_ADJ_MAX + 1) {
	111	lowmem_print(5, "lowmem_scan %lu, %x, return 0\n",
	112	sc->nr_to_scan, sc->gfp_mask);
	113	return 0;
355b0502	114	}
7dc19d5a	115
940f77b0	116	selected_oom_score_adj = min_score_adj;
355b0502	117
294b2711	118	rcu_read_lock();
95670001 AV	119	for_each_process(tsk) {
95670001 AV	120	struct task_struct *p;
a9c58b90	121	short oom_score_adj;
355b0502	122
9823ec9d	123	if (tsk->flags & PF_KTHREAD)
355b0502	124	continue;
9823ec9d	125
95670001 AV	126	p = find_lock_task_mm(tsk);
	127	if (!p)
	128	continue;
	129
83dbbdbb DR	130	if (test_tsk_thread_flag(p, TIF_MEMDIE) &&
	131	time_before_eq(jiffies, lowmem_deathpending_timeout)) {
	132	task_unlock(p);
	133	rcu_read_unlock();
	134	return 0;
	135	}
940f77b0 DR	136	oom_score_adj = p->signal->oom_score_adj;
940f77b0 DR	137	if (oom_score_adj < min_score_adj) {
355b0502 GKH	138	task_unlock(p);
	139	continue;
	140	}
95670001	141	tasksize = get_mm_rss(p->mm);
355b0502 GKH	142	task_unlock(p);
	143	if (tasksize <= 0)
	144	continue;
	145	if (selected) {
940f77b0	146	if (oom_score_adj < selected_oom_score_adj)
355b0502	147	continue;
940f77b0	148	if (oom_score_adj == selected_oom_score_adj &&
355b0502 GKH	149	tasksize <= selected_tasksize)
	150	continue;
	151	}
	152	selected = p;
	153	selected_tasksize = tasksize;
940f77b0	154	selected_oom_score_adj = oom_score_adj;
a9c58b90	155	lowmem_print(2, "select %d (%s), adj %hd, size %d, to kill\n",
940f77b0	156	p->pid, p->comm, oom_score_adj, tasksize);
355b0502 GKH	157	}
355b0502 GKH	158	if (selected) {
a9c58b90	159	lowmem_print(1, "send sigkill to %d (%s), adj %hd, size %d\n",
355b0502	160	selected->pid, selected->comm,
940f77b0	161	selected_oom_score_adj, selected_tasksize);
e5d7965f	162	lowmem_deathpending_timeout = jiffies + HZ;
49550b60 MH	163	/*
	164	* FIXME: lowmemorykiller shouldn't abuse global OOM killer
	165	* infrastructure. There is no real reason why the selected
	166	* task should have access to the memory reserves.
	167	*/
	168	mark_tsk_oom_victim(selected);
6bc2b856	169	send_sig(SIGKILL, selected, 0);
7dc19d5a	170	rem += selected_tasksize;
355b0502	171	}
7dc19d5a DC	172
7dc19d5a DC	173	lowmem_print(4, "lowmem_scan %lu, %x, return %lu\n",
cae9bf11	174	sc->nr_to_scan, sc->gfp_mask, rem);
294b2711	175	rcu_read_unlock();
355b0502 GKH	176	return rem;
	177	}
	178
	179	static struct shrinker lowmem_shrinker = {
7dc19d5a DC	180	.scan_objects = lowmem_scan,
7dc19d5a DC	181	.count_objects = lowmem_count,
355b0502 GKH	182	.seeks = DEFAULT_SEEKS * 16
	183	};
	184
	185	static int __init lowmem_init(void)
	186	{
	187	register_shrinker(&lowmem_shrinker);
	188	return 0;
	189	}
	190
	191	static void __exit lowmem_exit(void)
	192	{
	193	unregister_shrinker(&lowmem_shrinker);
	194	}
	195
	196	module_param_named(cost, lowmem_shrinker.seeks, int, S_IRUGO \| S_IWUSR);
a9c58b90	197	module_param_array_named(adj, lowmem_adj, short, &lowmem_adj_size,
355b0502 GKH	198	S_IRUGO \| S_IWUSR);
	199	module_param_array_named(minfree, lowmem_minfree, uint, &lowmem_minfree_size,
	200	S_IRUGO \| S_IWUSR);
	201	module_param_named(debug_level, lowmem_debug_level, uint, S_IRUGO \| S_IWUSR);
	202
	203	module_init(lowmem_init);
	204	module_exit(lowmem_exit);
	205
	206	MODULE_LICENSE("GPL");
	207