linux/mm/damon/core.c

// SPDX-License-Identifier: GPL-2.0
/*
 * Data Access Monitor
 *
 * Author: SeongJae Park <sj@kernel.org>
 */

#define pr_fmt(fmt) "damon: " fmt

#include <linux/damon.h>
#include <linux/delay.h>
#include <linux/kthread.h>
#include <linux/memcontrol.h>
#include <linux/mm.h>
#include <linux/psi.h>
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/string.h>
#include <linux/string_choices.h>

/* for damon_get_folio() used by node eligible memory metrics */
#include "ops-common.h"

#define CREATE_TRACE_POINTS
#include <trace/events/damon.h>

static DEFINE_MUTEX(damon_lock);
static int nr_running_ctxs;
static bool running_exclusive_ctxs;

static DEFINE_MUTEX(damon_ops_lock);
static struct damon_operations damon_registered_ops[NR_DAMON_OPS];

static struct kmem_cache *damon_region_cache __ro_after_init;

/* Should be called under damon_ops_lock with id smaller than NR_DAMON_OPS */
static bool __damon_is_registered_ops(enum damon_ops_id id)
{
	struct damon_operations empty_ops = {};

	if (!memcmp(&empty_ops, &damon_registered_ops[id], sizeof(empty_ops)))
		return false;
	return true;
}

/**
 * damon_is_registered_ops() - Check if a given damon_operations is registered.
 * @id:	Id of the damon_operations to check if registered.
 *
 * Return: true if the ops is set, false otherwise.
 */
bool damon_is_registered_ops(enum damon_ops_id id)
{
	bool registered;

	if (id >= NR_DAMON_OPS)
		return false;
	mutex_lock(&damon_ops_lock);
	registered = __damon_is_registered_ops(id);
	mutex_unlock(&damon_ops_lock);
	return registered;
}

/**
 * damon_register_ops() - Register a monitoring operations set to DAMON.
 * @ops:	monitoring operations set to register.
 *
 * This function registers a monitoring operations set of valid &struct
 * damon_operations->id so that others can find and use them later.
 *
 * Return: 0 on success, negative error code otherwise.
 */
int damon_register_ops(struct damon_operations *ops)
{
	int err = 0;

	if (ops->id >= NR_DAMON_OPS)
		return -EINVAL;

	mutex_lock(&damon_ops_lock);
	/* Fail for already registered ops */
	if (__damon_is_registered_ops(ops->id))
		err = -EINVAL;
	else
		damon_registered_ops[ops->id] = *ops;
	mutex_unlock(&damon_ops_lock);
	return err;
}

/**
 * damon_select_ops() - Select a monitoring operations to use with the context.
 * @ctx:	monitoring context to use the operations.
 * @id:		id of the registered monitoring operations to select.
 *
 * This function finds registered monitoring operations set of @id and make
 * @ctx to use it.
 *
 * Return: 0 on success, negative error code otherwise.
 */
int damon_select_ops(struct damon_ctx *ctx, enum damon_ops_id id)
{
	int err = 0;

	if (id >= NR_DAMON_OPS)
		return -EINVAL;

	mutex_lock(&damon_ops_lock);
	if (!__damon_is_registered_ops(id))
		err = -EINVAL;
	else
		ctx->ops = damon_registered_ops[id];
	mutex_unlock(&damon_ops_lock);
	return err;
}

struct damon_filter *damon_new_filter(enum damon_filter_type type,
		bool matching, bool allow)
{
	struct damon_filter *filter;

	filter = kmalloc_obj(*filter);
	if (!filter)
		return NULL;
	filter->type = type;
	filter->matching = matching;
	filter->allow = allow;
	INIT_LIST_HEAD(&filter->list);
	return filter;
}

void damon_add_filter(struct damon_probe *p, struct damon_filter *f)
{
	list_add_tail(&f->list, &p->filters);
}

static void damon_del_filter(struct damon_filter *f)
{
	list_del(&f->list);
}

static void damon_free_filter(struct damon_filter *f)
{
	kfree(f);
}

void damon_destroy_filter(struct damon_filter *f)
{
	damon_del_filter(f);
	damon_free_filter(f);
}

static struct damon_filter *damon_nth_filter(int n, struct damon_probe *p)
{
	struct damon_filter *f;
	int i = 0;

	damon_for_each_filter(f, p) {
		if (i++ == n)
			return f;
	}
	return NULL;
}

struct damon_probe *damon_new_probe(void)
{
	struct damon_probe *p;

	p = kmalloc_obj(*p);
	if (!p)
		return NULL;
	INIT_LIST_HEAD(&p->filters);
	INIT_LIST_HEAD(&p->list);
	return p;
}

void damon_add_probe(struct damon_ctx *ctx, struct damon_probe *probe)
{
	list_add_tail(&probe->list, &ctx->probes);
}

static void damon_del_probe(struct damon_probe *p)
{
	list_del(&p->list);
}

static void damon_free_probe(struct damon_probe *p)
{
	struct damon_filter *f, *next;

	damon_for_each_filter_safe(f, next, p)
		damon_free_filter(f);
	kfree(p);
}

static void damon_destroy_probe(struct damon_probe *p)
{
	damon_del_probe(p);
	damon_free_probe(p);
}

static struct damon_probe *damon_nth_probe(int n, struct damon_ctx *ctx)
{
	struct damon_probe *p;
	int i = 0;

	damon_for_each_probe(p, ctx) {
		if (i++ == n)
			return p;
	}
	return NULL;
}

#ifdef CONFIG_DAMON_DEBUG_SANITY
static void damon_verify_new_region(unsigned long start, unsigned long end)
{
	WARN_ONCE(start >= end, "start %lu >= end %lu\n", start, end);
}
#else
static void damon_verify_new_region(unsigned long start, unsigned long end)
{
}
#endif

/*
 * Construct a damon_region struct
 *
 * Returns the pointer to the new struct if success, or NULL otherwise
 */
struct damon_region *damon_new_region(unsigned long start, unsigned long end)
{
	struct damon_region *region;
	int i;

	damon_verify_new_region(start, end);
	region = kmem_cache_alloc(damon_region_cache, GFP_KERNEL);
	if (!region)
		return NULL;

	region->ar.start = start;
	region->ar.end = end;
	region->nr_accesses = 0;
	region->nr_accesses_bp = 0;
	for (i = 0; i < DAMON_MAX_PROBES; i++)
		region->probe_hits[i] = 0;
	INIT_LIST_HEAD(&region->list);

	region->age = 0;
	region->last_nr_accesses = 0;

	return region;
}

static void damon_add_region(struct damon_region *r, struct damon_target *t)
{
	list_add_tail(&r->list, &t->regions_list);
	t->nr_regions++;
}

/*
 * Add a region between two other regions
 */
static inline void damon_insert_region(struct damon_region *r,
		struct damon_region *prev, struct damon_region *next,
		struct damon_target *t)
{
	__list_add(&r->list, &prev->list, &next->list);
	t->nr_regions++;
}

#ifdef CONFIG_DAMON_DEBUG_SANITY
static void damon_verify_del_region(struct damon_target *t)
{
	WARN_ONCE(t->nr_regions == 0, "t->nr_regions == 0\n");
}
#else
static void damon_verify_del_region(struct damon_target *t)
{
}
#endif

static void damon_del_region(struct damon_region *r, struct damon_target *t)
{
	damon_verify_del_region(t);

	list_del(&r->list);
	t->nr_regions--;
}

static void damon_free_region(struct damon_region *r)
{
	kmem_cache_free(damon_region_cache, r);
}

static void damon_destroy_region(struct damon_region *r,
		struct damon_target *t)
{
	damon_del_region(r, t);
	damon_free_region(r);
}

static bool damon_is_last_region(struct damon_region *r,
		struct damon_target *t)
{
	return list_is_last(&r->list, &t->regions_list);
}

/*
 * Check whether a region is intersecting an address range
 *
 * Returns true if it is.
 */
static bool damon_intersect(struct damon_region *r,
		struct damon_addr_range *re)
{
	return !(r->ar.end <= re->start || re->end <= r->ar.start);
}

/*
 * Fill holes in regions with new regions.
 */
static int damon_fill_regions_holes(struct damon_region *first,
		struct damon_region *last, struct damon_target *t)
{
	struct damon_region *r = first;

	damon_for_each_region_from(r, t) {
		struct damon_region *next, *newr;

		if (r == last)
			break;
		next = damon_next_region(r);
		if (r->ar.end != next->ar.start) {
			newr = damon_new_region(r->ar.end, next->ar.start);
			if (!newr)
				return -ENOMEM;
			damon_insert_region(newr, r, next, t);
		}
	}
	return 0;
}

/*
 * damon_set_regions() - Set regions of a target for given address ranges.
 * @t:		the given target.
 * @ranges:	array of new monitoring target ranges.
 * @nr_ranges:	length of @ranges.
 * @min_region_sz:	minimum region size.
 *
 * This function adds new regions to, or modify existing regions of a
 * monitoring target to fit in specific ranges.
 *
 * Return: 0 if success, or negative error code otherwise.
 */
int damon_set_regions(struct damon_target *t, struct damon_addr_range *ranges,
		unsigned int nr_ranges, unsigned long min_region_sz)
{
	struct damon_region *r, *next;
	unsigned int i;
	int err;

	/* Remove regions which are not in the new ranges */
	damon_for_each_region_safe(r, next, t) {
		for (i = 0; i < nr_ranges; i++) {
			if (damon_intersect(r, &ranges[i]))
				break;
		}
		if (i == nr_ranges)
			damon_destroy_region(r, t);
	}

	if (!damon_nr_regions(t)) {
		for (i = 0; i < nr_ranges; i++) {
			r = damon_new_region(
					ALIGN_DOWN(ranges[i].start,
						min_region_sz),
					ALIGN(ranges[i].end, min_region_sz));
			if (!r)
				return -ENOMEM;
			damon_add_region(r, t);
		}
		return 0;
	}

	r = damon_first_region(t);
	/* Add new regions or resize existing regions to fit in the ranges */
	for (i = 0; i < nr_ranges; i++) {
		struct damon_region *first = NULL, *last, *newr;
		struct damon_addr_range *range;
		bool insert_before_r = false;

		range = &ranges[i];
		/* Get the first/last regions intersecting with the range */
		damon_for_each_region_from(r, t) {
			if (damon_intersect(r, range)) {
				if (!first)
					first = r;
				last = r;
			}
			if (r->ar.start >= range->end) {
				insert_before_r = true;
				break;
			}
		}
		if (!first) {
			/* no region intersects with this range */
			newr = damon_new_region(
					ALIGN_DOWN(range->start,
						min_region_sz),
					ALIGN(range->end, min_region_sz));
			if (!newr)
				return -ENOMEM;
			if (insert_before_r)
				damon_insert_region(newr, damon_prev_region(r),
						r, t);
			else
				damon_add_region(newr, t);
		} else {
			/* resize intersecting regions to fit in this range */
			first->ar.start = ALIGN_DOWN(range->start,
					min_region_sz);
			last->ar.end = ALIGN(range->end, min_region_sz);

			/* fill possible holes in the range */
			err = damon_fill_regions_holes(first, last, t);
			if (err)
				return err;
		}
	}
	return 0;
}

struct damos_filter *damos_new_filter(enum damos_filter_type type,
		bool matching, bool allow)
{
	struct damos_filter *filter;

	filter = kmalloc_obj(*filter);
	if (!filter)
		return NULL;
	filter->type = type;
	filter->matching = matching;
	filter->allow = allow;
	INIT_LIST_HEAD(&filter->list);
	return filter;
}

/**
 * damos_filter_for_ops() - Return if the filter is ops-handled one.
 * @type:	type of the filter.
 *
 * Return: true if the filter of @type needs to be handled by ops layer, false
 * otherwise.
 */
bool damos_filter_for_ops(enum damos_filter_type type)
{
	switch (type) {
	case DAMOS_FILTER_TYPE_ADDR:
	case DAMOS_FILTER_TYPE_TARGET:
		return false;
	default:
		break;
	}
	return true;
}

void damos_add_filter(struct damos *s, struct damos_filter *f)
{
	if (damos_filter_for_ops(f->type))
		list_add_tail(&f->list, &s->ops_filters);
	else
		list_add_tail(&f->list, &s->core_filters);
}

static void damos_del_filter(struct damos_filter *f)
{
	list_del(&f->list);
}

static void damos_free_filter(struct damos_filter *f)
{
	kfree(f);
}

void damos_destroy_filter(struct damos_filter *f)
{
	damos_del_filter(f);
	damos_free_filter(f);
}

struct damos_quota_goal *damos_new_quota_goal(
		enum damos_quota_goal_metric metric,
		unsigned long target_value)
{
	struct damos_quota_goal *goal;

	goal = kmalloc_obj(*goal);
	if (!goal)
		return NULL;
	goal->metric = metric;
	goal->target_value = target_value;
	INIT_LIST_HEAD(&goal->list);
	return goal;
}

void damos_add_quota_goal(struct damos_quota *q, struct damos_quota_goal *g)
{
	list_add_tail(&g->list, &q->goals);
}

static void damos_del_quota_goal(struct damos_quota_goal *g)
{
	list_del(&g->list);
}

static void damos_free_quota_goal(struct damos_quota_goal *g)
{
	kfree(g);
}

void damos_destroy_quota_goal(struct damos_quota_goal *g)
{
	damos_del_quota_goal(g);
	damos_free_quota_goal(g);
}

static bool damos_quota_goals_empty(struct damos_quota *q)
{
	return list_empty(&q->goals);
}

/* initialize fields of @quota that normally API users wouldn't set */
static struct damos_quota *damos_quota_init(struct damos_quota *quota)
{
	quota->esz = 0;
	quota->total_charged_sz = 0;
	quota->total_charged_ns = 0;
	quota->charged_sz = 0;
	quota->charged_from = 0;
	quota->charge_target_from = NULL;
	quota->charge_addr_from = 0;
	quota->esz_bp = 0;
	return quota;
}

struct damos *damon_new_scheme(struct damos_access_pattern *pattern,
			enum damos_action action,
			unsigned long apply_interval_us,
			struct damos_quota *quota,
			struct damos_watermarks *wmarks,
			int target_nid)
{
	struct damos *scheme;

	scheme = kmalloc_obj(*scheme);
	if (!scheme)
		return NULL;
	scheme->pattern = *pattern;
	scheme->action = action;
	scheme->apply_interval_us = apply_interval_us;
	/*
	 * next_apply_sis will be set when kdamond starts.  While kdamond is
	 * running, it will also updated when it is added to the DAMON context,
	 * or damon_attrs are updated.
	 */
	scheme->next_apply_sis = 0;
	scheme->walk_completed = false;
	INIT_LIST_HEAD(&scheme->core_filters);
	INIT_LIST_HEAD(&scheme->ops_filters);
	scheme->stat = (struct damos_stat){};
	scheme->max_nr_snapshots = 0;
	INIT_LIST_HEAD(&scheme->list);

	scheme->quota = *(damos_quota_init(quota));
	/* quota.goals should be separately set by caller */
	INIT_LIST_HEAD(&scheme->quota.goals);

	scheme->wmarks = *wmarks;
	scheme->wmarks.activated = true;

	scheme->migrate_dests = (struct damos_migrate_dests){};
	scheme->target_nid = target_nid;

	return scheme;
}

static void damos_set_next_apply_sis(struct damos *s, struct damon_ctx *ctx)
{
	unsigned long sample_interval = ctx->attrs.sample_interval ?
		ctx->attrs.sample_interval : 1;
	unsigned long apply_interval = s->apply_interval_us ?
		s->apply_interval_us : ctx->attrs.aggr_interval;

	s->next_apply_sis = ctx->passed_sample_intervals +
		apply_interval / sample_interval;
}

void damon_add_scheme(struct damon_ctx *ctx, struct damos *s)
{
	list_add_tail(&s->list, &ctx->schemes);
	damos_set_next_apply_sis(s, ctx);
}

static void damon_del_scheme(struct damos *s)
{
	list_del(&s->list);
}

static void damon_free_scheme(struct damos *s)
{
	kfree(s);
}

void damon_destroy_scheme(struct damos *s)
{
	struct damos_quota_goal *g, *g_next;
	struct damos_filter *f, *next;

	damos_for_each_quota_goal_safe(g, g_next, &s->quota)
		damos_destroy_quota_goal(g);

	damos_for_each_core_filter_safe(f, next, s)
		damos_destroy_filter(f);

	damos_for_each_ops_filter_safe(f, next, s)
		damos_destroy_filter(f);

	kfree(s->migrate_dests.node_id_arr);
	kfree(s->migrate_dests.weight_arr);
	damon_del_scheme(s);
	damon_free_scheme(s);
}

/*
 * Construct a damon_target struct
 *
 * Returns the pointer to the new struct if success, or NULL otherwise
 */
struct damon_target *damon_new_target(void)
{
	struct damon_target *t;

	t = kmalloc_obj(*t);
	if (!t)
		return NULL;

	t->pid = NULL;
	t->nr_regions = 0;
	INIT_LIST_HEAD(&t->regions_list);
	INIT_LIST_HEAD(&t->list);
	t->obsolete = false;

	return t;
}

void damon_add_target(struct damon_ctx *ctx, struct damon_target *t)
{
	list_add_tail(&t->list, &ctx->adaptive_targets);
}

bool damon_targets_empty(struct damon_ctx *ctx)
{
	return list_empty(&ctx->adaptive_targets);
}

static void damon_del_target(struct damon_target *t)
{
	list_del(&t->list);
}

void damon_free_target(struct damon_target *t)
{
	struct damon_region *r, *next;

	damon_for_each_region_safe(r, next, t)
		damon_free_region(r);
	kfree(t);
}

void damon_destroy_target(struct damon_target *t, struct damon_ctx *ctx)
{

	if (ctx && ctx->ops.cleanup_target)
		ctx->ops.cleanup_target(t);

	damon_del_target(t);
	damon_free_target(t);
}

unsigned int damon_nr_regions(struct damon_target *t)
{
	return t->nr_regions;
}

struct damon_ctx *damon_new_ctx(void)
{
	struct damon_ctx *ctx;

	ctx = kzalloc_obj(*ctx);
	if (!ctx)
		return NULL;

	init_completion(&ctx->kdamond_started);

	ctx->attrs.sample_interval = 5 * 1000;
	ctx->attrs.aggr_interval = 100 * 1000;
	ctx->attrs.ops_update_interval = 60 * 1000 * 1000;

	ctx->passed_sample_intervals = 0;
	/* These will be set from kdamond_init_ctx() */
	ctx->next_aggregation_sis = 0;
	ctx->next_ops_update_sis = 0;

	mutex_init(&ctx->kdamond_lock);
	INIT_LIST_HEAD(&ctx->call_controls);
	mutex_init(&ctx->call_controls_lock);
	mutex_init(&ctx->walk_control_lock);

	ctx->attrs.min_nr_regions = 10;
	ctx->attrs.max_nr_regions = 1000;

	INIT_LIST_HEAD(&ctx->probes);

	ctx->addr_unit = 1;
	ctx->min_region_sz = DAMON_MIN_REGION_SZ;

	INIT_LIST_HEAD(&ctx->adaptive_targets);
	INIT_LIST_HEAD(&ctx->schemes);

	prandom_seed_state(&ctx->rnd_state, get_random_u64());

	return ctx;
}

static void damon_destroy_targets(struct damon_ctx *ctx)
{
	struct damon_target *t, *next_t;

	damon_for_each_target_safe(t, next_t, ctx)
		damon_destroy_target(t, ctx);
}

void damon_destroy_ctx(struct damon_ctx *ctx)
{
	struct damos *s, *next_s;
	struct damon_probe *p, *next_p;

	damon_destroy_targets(ctx);

	damon_for_each_scheme_safe(s, next_s, ctx)
		damon_destroy_scheme(s);

	damon_for_each_probe_safe(p, next_p, ctx)
		damon_destroy_probe(p);

	kfree(ctx);
}

static bool damon_attrs_equals(const struct damon_attrs *attrs1,
		const struct damon_attrs *attrs2)
{
	const struct damon_intervals_goal *ig1 = &attrs1->intervals_goal;
	const struct damon_intervals_goal *ig2 = &attrs2->intervals_goal;

	return attrs1->sample_interval == attrs2->sample_interval &&
		attrs1->aggr_interval == attrs2->aggr_interval &&
		attrs1->ops_update_interval == attrs2->ops_update_interval &&
		attrs1->min_nr_regions == attrs2->min_nr_regions &&
		attrs1->max_nr_regions == attrs2->max_nr_regions &&
		ig1->access_bp == ig2->access_bp &&
		ig1->aggrs == ig2->aggrs &&
		ig1->min_sample_us == ig2->min_sample_us &&
		ig1->max_sample_us == ig2->max_sample_us;
}

static unsigned int damon_age_for_new_attrs(unsigned int age,
		struct damon_attrs *old_attrs, struct damon_attrs *new_attrs)
{
	return age * old_attrs->aggr_interval / new_attrs->aggr_interval;
}

/* convert access ratio in bp (per 10,000) to nr_accesses */
static unsigned int damon_accesses_bp_to_nr_accesses(
		unsigned int accesses_bp, struct damon_attrs *attrs)
{
	return accesses_bp * damon_max_nr_accesses(attrs) / 10000;
}

/*
 * Convert nr_accesses to access ratio in bp (per 10,000).
 *
 * Callers should ensure attrs.aggr_interval is not zero, like
 * damon_update_monitoring_results() does .  Otherwise, divide-by-zero would
 * happen.
 */
static unsigned int damon_nr_accesses_to_accesses_bp(
		unsigned int nr_accesses, struct damon_attrs *attrs)
{
	return mult_frac(nr_accesses, 10000, damon_max_nr_accesses(attrs));
}

static unsigned int damon_nr_accesses_for_new_attrs(unsigned int nr_accesses,
		struct damon_attrs *old_attrs, struct damon_attrs *new_attrs)
{
	return damon_accesses_bp_to_nr_accesses(
			damon_nr_accesses_to_accesses_bp(
				nr_accesses, old_attrs),
			new_attrs);
}

static void damon_update_monitoring_result(struct damon_region *r,
		struct damon_attrs *old_attrs, struct damon_attrs *new_attrs,
		bool aggregating)
{
	if (!aggregating) {
		r->nr_accesses = damon_nr_accesses_for_new_attrs(
				r->nr_accesses, old_attrs, new_attrs);
		r->nr_accesses_bp = r->nr_accesses * 10000;
	} else {
		/*
		 * if this is called in the middle of the aggregation, reset
		 * the aggregations we made so far for this aggregation
		 * interval.  In other words, make the status like
		 * kdamond_reset_aggregated() is called.
		 */
		r->last_nr_accesses = damon_nr_accesses_for_new_attrs(
				r->last_nr_accesses, old_attrs, new_attrs);
		r->nr_accesses_bp = r->last_nr_accesses * 10000;
		r->nr_accesses = 0;
	}
	r->age = damon_age_for_new_attrs(r->age, old_attrs, new_attrs);
}

/*
 * region->nr_accesses is the number of sampling intervals in the last
 * aggregation interval that access to the region has found, and region->age is
 * the number of aggregation intervals that its access pattern has maintained.
 * For the reason, the real meaning of the two fields depend on current
 * sampling interval and aggregation interval.  This function updates
 * ->nr_accesses and ->age of given damon_ctx's regions for new damon_attrs.
 */
static void damon_update_monitoring_results(struct damon_ctx *ctx,
		struct damon_attrs *new_attrs, bool aggregating)
{
	struct damon_attrs *old_attrs = &ctx->attrs;
	struct damon_target *t;
	struct damon_region *r;

	/* if any interval is zero, simply forgive conversion */
	if (!old_attrs->sample_interval || !old_attrs->aggr_interval ||
			!new_attrs->sample_interval ||
			!new_attrs->aggr_interval)
		return;

	damon_for_each_target(t, ctx)
		damon_for_each_region(r, t)
			damon_update_monitoring_result(
					r, old_attrs, new_attrs, aggregating);
}

/*
 * damon_valid_intervals_goal() - return if the intervals goal of @attrs is
 * valid.
 */
static bool damon_valid_intervals_goal(struct damon_attrs *attrs)
{
	struct damon_intervals_goal *goal = &attrs->intervals_goal;

	/* tuning is disabled */
	if (!goal->aggrs)
		return true;
	if (goal->min_sample_us > goal->max_sample_us)
		return false;
	if (attrs->sample_interval < goal->min_sample_us ||
			goal->max_sample_us < attrs->sample_interval)
		return false;
	return true;
}

/**
 * damon_set_attrs() - Set attributes for the monitoring.
 * @ctx:		monitoring context
 * @attrs:		monitoring attributes
 *
 * This function updates monitoring results and next monitoring/damos operation
 * schedules.  Because those are periodically updated by kdamond, this should
 * be called from a safe contexts.  Such contexts include damon_ctx setup time
 * while the kdamond is not yet started, and inside of kdamond_fn().
 *
 * In detail, all DAMON API callers directly call this function for initial
 * setup of damon_ctx before calling damon_start().  Some of the API callers
 * also indirectly call this function via damon_call() -> damon_commit() for
 * online parameters updates.  Finally, kdamond_fn() itself use this for
 * applying auto-tuned monitoring intervals.
 *
 * Every time interval is in micro-seconds.
 *
 * Return: 0 on success, negative error code otherwise.
 */
int damon_set_attrs(struct damon_ctx *ctx, struct damon_attrs *attrs)
{
	unsigned long sample_interval = attrs->sample_interval ?
		attrs->sample_interval : 1;
	struct damos *s;
	bool aggregating = ctx->passed_sample_intervals <
		ctx->next_aggregation_sis;

	if (!damon_valid_intervals_goal(attrs))
		return -EINVAL;

	if (attrs->min_nr_regions < 3)
		return -EINVAL;
	if (attrs->min_nr_regions > attrs->max_nr_regions)
		return -EINVAL;
	if (attrs->sample_interval > attrs->aggr_interval)
		return -EINVAL;

	/* calls from core-external doesn't set this. */
	if (!attrs->aggr_samples)
		attrs->aggr_samples = attrs->aggr_interval / sample_interval;

	ctx->next_aggregation_sis = ctx->passed_sample_intervals +
		attrs->aggr_interval / sample_interval;
	ctx->next_ops_update_sis = ctx->passed_sample_intervals +
		attrs->ops_update_interval / sample_interval;
	/*
	 * next_intervals_tune_sis will be updated inside kdamond_fn().
	 */

	damon_update_monitoring_results(ctx, attrs, aggregating);
	ctx->attrs = *attrs;

	damon_for_each_scheme(s, ctx)
		damos_set_next_apply_sis(s, ctx);

	return 0;
}

/**
 * damon_set_schemes() - Set data access monitoring based operation schemes.
 * @ctx:	monitoring context
 * @schemes:	array of the schemes
 * @nr_schemes:	number of entries in @schemes
 *
 * This function should not be called while the kdamond of the context is
 * running.
 */
void damon_set_schemes(struct damon_ctx *ctx, struct damos **schemes,
			ssize_t nr_schemes)
{
	struct damos *s, *next;
	ssize_t i;

	damon_for_each_scheme_safe(s, next, ctx)
		damon_destroy_scheme(s);
	for (i = 0; i < nr_schemes; i++)
		damon_add_scheme(ctx, schemes[i]);
}

static struct damos_quota_goal *damos_nth_quota_goal(
		int n, struct damos_quota *q)
{
	struct damos_quota_goal *goal;
	int i = 0;

	damos_for_each_quota_goal(goal, q) {
		if (i++ == n)
			return goal;
	}
	return NULL;
}

static void damos_commit_quota_goal_union(
		struct damos_quota_goal *dst, struct damos_quota_goal *src)
{
	switch (dst->metric) {
	case DAMOS_QUOTA_NODE_MEM_USED_BP:
	case DAMOS_QUOTA_NODE_MEM_FREE_BP:
		dst->nid = src->nid;
		break;
	case DAMOS_QUOTA_NODE_MEMCG_USED_BP:
	case DAMOS_QUOTA_NODE_MEMCG_FREE_BP:
		dst->nid = src->nid;
		dst->memcg_id = src->memcg_id;
		break;
	default:
		break;
	}
}

static void damos_commit_quota_goal(
		struct damos_quota_goal *dst, struct damos_quota_goal *src)
{
	dst->metric = src->metric;
	dst->target_value = src->target_value;
	if (dst->metric == DAMOS_QUOTA_USER_INPUT)
		dst->current_value = src->current_value;
	/* keep last_psi_total as is, since it will be updated in next cycle */
	damos_commit_quota_goal_union(dst, src);
}

/**
 * damos_commit_quota_goals() - Commit DAMOS quota goals to another quota.
 * @dst:	The commit destination DAMOS quota.
 * @src:	The commit source DAMOS quota.
 *
 * Copies user-specified parameters for quota goals from @src to @dst.  Users
 * should use this function for quota goals-level parameters update of running
 * DAMON contexts, instead of manual in-place updates.
 *
 * This function should be called from parameters-update safe context, like
 * damon_call().
 */
int damos_commit_quota_goals(struct damos_quota *dst, struct damos_quota *src)
{
	struct damos_quota_goal *dst_goal, *next, *src_goal, *new_goal;
	int i = 0, j = 0;

	damos_for_each_quota_goal_safe(dst_goal, next, dst) {
		src_goal = damos_nth_quota_goal(i++, src);
		if (src_goal)
			damos_commit_quota_goal(dst_goal, src_goal);
		else
			damos_destroy_quota_goal(dst_goal);
	}
	damos_for_each_quota_goal_safe(src_goal, next, src) {
		if (j++ < i)
			continue;
		new_goal = damos_new_quota_goal(
				src_goal->metric, src_goal->target_value);
		if (!new_goal)
			return -ENOMEM;
		damos_commit_quota_goal(new_goal, src_goal);
		damos_add_quota_goal(dst, new_goal);
	}
	return 0;
}

static int damos_commit_quota(struct damos_quota *dst, struct damos_quota *src)
{
	int err;

	dst->reset_interval = src->reset_interval;
	dst->ms = src->ms;
	dst->sz = src->sz;
	err = damos_commit_quota_goals(dst, src);
	if (err)
		return err;
	dst->goal_tuner = src->goal_tuner;
	dst->fail_charge_num = src->fail_charge_num;
	dst->fail_charge_denom = src->fail_charge_denom;
	dst->weight_sz = src->weight_sz;
	dst->weight_nr_accesses = src->weight_nr_accesses;
	dst->weight_age = src->weight_age;
	return 0;
}

static struct damos_filter *damos_nth_core_filter(int n, struct damos *s)
{
	struct damos_filter *filter;
	int i = 0;

	damos_for_each_core_filter(filter, s) {
		if (i++ == n)
			return filter;
	}
	return NULL;
}

static struct damos_filter *damos_nth_ops_filter(int n, struct damos *s)
{
	struct damos_filter *filter;
	int i = 0;

	damos_for_each_ops_filter(filter, s) {
		if (i++ == n)
			return filter;
	}
	return NULL;
}

static void damos_commit_filter_arg(
		struct damos_filter *dst, struct damos_filter *src)
{
	switch (dst->type) {
	case DAMOS_FILTER_TYPE_MEMCG:
		dst->memcg_id = src->memcg_id;
		break;
	case DAMOS_FILTER_TYPE_ADDR:
		dst->addr_range = src->addr_range;
		break;
	case DAMOS_FILTER_TYPE_TARGET:
		dst->target_idx = src->target_idx;
		break;
	case DAMOS_FILTER_TYPE_HUGEPAGE_SIZE:
		dst->sz_range = src->sz_range;
		break;
	default:
		break;
	}
}

static void damos_commit_filter(
		struct damos_filter *dst, struct damos_filter *src)
{
	dst->type = src->type;
	dst->matching = src->matching;
	dst->allow = src->allow;
	damos_commit_filter_arg(dst, src);
}

static int damos_commit_core_filters(struct damos *dst, struct damos *src)
{
	struct damos_filter *dst_filter, *next, *src_filter, *new_filter;
	int i = 0, j = 0;

	damos_for_each_core_filter_safe(dst_filter, next, dst) {
		src_filter = damos_nth_core_filter(i++, src);
		if (src_filter)
			damos_commit_filter(dst_filter, src_filter);
		else
			damos_destroy_filter(dst_filter);
	}

	damos_for_each_core_filter_safe(src_filter, next, src) {
		if (j++ < i)
			continue;

		new_filter = damos_new_filter(
				src_filter->type, src_filter->matching,
				src_filter->allow);
		if (!new_filter)
			return -ENOMEM;
		damos_commit_filter_arg(new_filter, src_filter);
		damos_add_filter(dst, new_filter);
	}
	return 0;
}

static int damos_commit_ops_filters(struct damos *dst, struct damos *src)
{
	struct damos_filter *dst_filter, *next, *src_filter, *new_filter;
	int i = 0, j = 0;

	damos_for_each_ops_filter_safe(dst_filter, next, dst) {
		src_filter = damos_nth_ops_filter(i++, src);
		if (src_filter)
			damos_commit_filter(dst_filter, src_filter);
		else
			damos_destroy_filter(dst_filter);
	}

	damos_for_each_ops_filter_safe(src_filter, next, src) {
		if (j++ < i)
			continue;

		new_filter = damos_new_filter(
				src_filter->type, src_filter->matching,
				src_filter->allow);
		if (!new_filter)
			return -ENOMEM;
		damos_commit_filter_arg(new_filter, src_filter);
		damos_add_filter(dst, new_filter);
	}
	return 0;
}

/**
 * damos_filters_default_reject() - decide whether to reject memory that didn't
 *				    match with any given filter.
 * @filters:	Given DAMOS filters of a group.
 */
static bool damos_filters_default_reject(struct list_head *filters)
{
	struct damos_filter *last_filter;

	if (list_empty(filters))
		return false;
	last_filter = list_last_entry(filters, struct damos_filter, list);
	return last_filter->allow;
}

static void damos_set_filters_default_reject(struct damos *s)
{
	if (!list_empty(&s->ops_filters))
		s->core_filters_default_reject = false;
	else
		s->core_filters_default_reject =
			damos_filters_default_reject(&s->core_filters);
	s->ops_filters_default_reject =
		damos_filters_default_reject(&s->ops_filters);
}

/*
 * damos_commit_dests() - Copy migration destinations from @src to @dst.
 * @dst:	Destination structure to update.
 * @src:	Source structure to copy from.
 *
 * If the number of destinations has changed, the old arrays in @dst are freed
 * and new ones are allocated.  On success, @dst contains a full copy of
 * @src's arrays and count.
 *
 * On allocation failure, @dst is left in a partially torn-down state: its
 * arrays may be NULL and @nr_dests may not reflect the actual allocation
 * sizes.  The structure remains safe to deallocate via damon_destroy_scheme(),
 * but callers must not reuse @dst for further commits — it should be
 * discarded.
 *
 * Return: 0 on success, -ENOMEM on allocation failure.
 */
static int damos_commit_dests(struct damos_migrate_dests *dst,
		struct damos_migrate_dests *src)
{
	if (dst->nr_dests != src->nr_dests) {
		kfree(dst->node_id_arr);
		kfree(dst->weight_arr);

		dst->node_id_arr = kmalloc_array(src->nr_dests,
			sizeof(*dst->node_id_arr), GFP_KERNEL);
		if (!dst->node_id_arr) {
			dst->weight_arr = NULL;
			return -ENOMEM;
		}

		dst->weight_arr = kmalloc_array(src->nr_dests,
			sizeof(*dst->weight_arr), GFP_KERNEL);
		if (!dst->weight_arr) {
			/* ->node_id_arr will be freed by scheme destruction */
			return -ENOMEM;
		}
	}

	dst->nr_dests = src->nr_dests;
	for (int i = 0; i < src->nr_dests; i++) {
		dst->node_id_arr[i] = src->node_id_arr[i];
		dst->weight_arr[i] = src->weight_arr[i];
	}

	return 0;
}

static int damos_commit_filters(struct damos *dst, struct damos *src)
{
	int err;

	err = damos_commit_core_filters(dst, src);
	if (err)
		return err;
	err = damos_commit_ops_filters(dst, src);
	if (err)
		return err;
	damos_set_filters_default_reject(dst);
	return 0;
}

static struct damos *damon_nth_scheme(int n, struct damon_ctx *ctx)
{
	struct damos *s;
	int i = 0;

	damon_for_each_scheme(s, ctx) {
		if (i++ == n)
			return s;
	}
	return NULL;
}

static int damos_commit(struct damos *dst, struct damos *src)
{
	int err;

	dst->pattern = src->pattern;
	dst->action = src->action;
	dst->apply_interval_us = src->apply_interval_us;

	err = damos_commit_quota(&dst->quota, &src->quota);
	if (err)
		return err;

	dst->wmarks = src->wmarks;
	dst->target_nid = src->target_nid;

	err = damos_commit_dests(&dst->migrate_dests, &src->migrate_dests);
	if (err)
		return err;

	err = damos_commit_filters(dst, src);
	if (err)
		return err;

	dst->max_nr_snapshots = src->max_nr_snapshots;
	return 0;
}

static int damon_commit_schemes(struct damon_ctx *dst, struct damon_ctx *src)
{
	struct damos *dst_scheme, *next, *src_scheme, *new_scheme;
	int i = 0, j = 0, err;

	damon_for_each_scheme_safe(dst_scheme, next, dst) {
		src_scheme = damon_nth_scheme(i++, src);
		if (src_scheme) {
			err = damos_commit(dst_scheme, src_scheme);
			if (err)
				return err;
		} else {
			damon_destroy_scheme(dst_scheme);
		}
	}

	damon_for_each_scheme_safe(src_scheme, next, src) {
		if (j++ < i)
			continue;
		new_scheme = damon_new_scheme(&src_scheme->pattern,
				src_scheme->action,
				src_scheme->apply_interval_us,
				&src_scheme->quota, &src_scheme->wmarks,
				NUMA_NO_NODE);
		if (!new_scheme)
			return -ENOMEM;
		err = damos_commit(new_scheme, src_scheme);
		if (err) {
			damon_destroy_scheme(new_scheme);
			return err;
		}
		damon_add_scheme(dst, new_scheme);
	}
	return 0;
}

static struct damon_target *damon_nth_target(int n, struct damon_ctx *ctx)
{
	struct damon_target *t;
	int i = 0;

	damon_for_each_target(t, ctx) {
		if (i++ == n)
			return t;
	}
	return NULL;
}

/*
 * The caller should ensure the regions of @src are
 * 1. valid (end >= src) and
 * 2. sorted by starting address.
 *
 * If @src has no region, @dst keeps current regions.
 */
static int damon_commit_target_regions(struct damon_target *dst,
		struct damon_target *src, unsigned long src_min_region_sz)
{
	struct damon_region *src_region;
	struct damon_addr_range *ranges;
	int i = 0, err;

	damon_for_each_region(src_region, src)
		i++;
	if (!i)
		return 0;

	ranges = kmalloc_objs(*ranges, i, GFP_KERNEL | __GFP_NOWARN);
	if (!ranges)
		return -ENOMEM;
	i = 0;
	damon_for_each_region(src_region, src)
		ranges[i++] = src_region->ar;
	err = damon_set_regions(dst, ranges, i, src_min_region_sz);
	kfree(ranges);
	return err;
}

static int damon_commit_target(
		struct damon_target *dst, bool dst_has_pid,
		struct damon_target *src, bool src_has_pid,
		unsigned long src_min_region_sz)
{
	int err;

	err = damon_commit_target_regions(dst, src, src_min_region_sz);
	if (err)
		return err;
	if (dst_has_pid)
		put_pid(dst->pid);
	if (src_has_pid)
		get_pid(src->pid);
	dst->pid = src->pid;
	return 0;
}

static int damon_commit_targets(
		struct damon_ctx *dst, struct damon_ctx *src)
{
	struct damon_target *dst_target, *next, *src_target, *new_target;
	int i = 0, j = 0, err;

	damon_for_each_target_safe(dst_target, next, dst) {
		src_target = damon_nth_target(i++, src);
		/*
		 * If src target is obsolete, do not commit the parameters to
		 * the dst target, and further remove the dst target.
		 */
		if (src_target && !src_target->obsolete) {
			err = damon_commit_target(
					dst_target, damon_target_has_pid(dst),
					src_target, damon_target_has_pid(src),
					src->min_region_sz);
			if (err)
				return err;
		} else {
			struct damos *s;

			damon_destroy_target(dst_target, dst);
			damon_for_each_scheme(s, dst) {
				if (s->quota.charge_target_from == dst_target) {
					s->quota.charge_target_from = NULL;
					s->quota.charge_addr_from = 0;
				}
			}
		}
	}

	damon_for_each_target_safe(src_target, next, src) {
		if (j++ < i)
			continue;
		/* target to remove has no matching dst */
		if (src_target->obsolete)
			return -EINVAL;
		new_target = damon_new_target();
		if (!new_target)
			return -ENOMEM;
		err = damon_commit_target(new_target, false,
				src_target, damon_target_has_pid(src),
				src->min_region_sz);
		if (err) {
			damon_destroy_target(new_target, NULL);
			return err;
		}
		damon_add_target(dst, new_target);
	}
	return 0;
}

static void damon_commit_filter(struct damon_filter *dst,
		struct damon_filter *src)
{
	dst->type = src->type;
	dst->matching = src->matching;
	dst->allow = src->allow;
	switch (dst->type) {
	case DAMON_FILTER_TYPE_MEMCG:
		dst->memcg_id = src->memcg_id;
		break;
	default:
		break;
	}
}

static int damon_commit_filters(struct damon_probe *dst,
		struct damon_probe *src)
{
	struct damon_filter *dst_filter, *next, *src_filter, *new_filter;
	int i = 0, j = 0;

	damon_for_each_filter_safe(dst_filter, next, dst) {
		src_filter = damon_nth_filter(i++, src);
		if (src_filter)
			damon_commit_filter(dst_filter, src_filter);
		else
			damon_destroy_filter(dst_filter);
	}

	damon_for_each_filter_safe(src_filter, next, src) {
		if (j++ < i)
			continue;

		new_filter = damon_new_filter(src_filter->type,
				src_filter->matching, src_filter->allow);
		if (!new_filter)
			return -ENOMEM;
		switch (src_filter->type) {
		case DAMON_FILTER_TYPE_MEMCG:
			new_filter->memcg_id = src_filter->memcg_id;
			break;
		default:
			break;
		}
		damon_add_filter(dst, new_filter);
	}
	return 0;
}

static int damon_commit_probes(struct damon_ctx *dst, struct damon_ctx *src)
{
	struct damon_probe *dst_probe, *next, *src_probe, *new_probe;
	int i = 0, j = 0, err;

	damon_for_each_probe_safe(dst_probe, next, dst) {
		src_probe = damon_nth_probe(i++, src);
		if (src_probe) {
			err = damon_commit_filters(dst_probe, src_probe);
			if (err)
				return err;
		} else {
			damon_destroy_probe(dst_probe);
		}
	}

	damon_for_each_probe_safe(src_probe, next, src) {
		if (j++ < i)
			continue;

		new_probe = damon_new_probe();
		if (!new_probe)
			return -ENOMEM;
		damon_add_probe(dst, new_probe);
		err = damon_commit_filters(new_probe, src_probe);
		if (err)
			return err;
	}
	return 0;
}

/**
 * damon_commit_ctx() - Commit parameters of a DAMON context to another.
 * @dst:	The commit destination DAMON context.
 * @src:	The commit source DAMON context.
 *
 * This function copies user-specified parameters from @src to @dst and update
 * the internal status and results accordingly.  Users should use this function
 * for context-level parameters update of running context, instead of manual
 * in-place updates.
 *
 * This function should be called from parameters-update safe context, like
 * damon_call().
 */
int damon_commit_ctx(struct damon_ctx *dst, struct damon_ctx *src)
{
	int err;
	struct damos *scheme;
	struct damos_quota_goal *goal;

	dst->maybe_corrupted = true;
	if (!is_power_of_2(src->min_region_sz))
		return -EINVAL;

	/* node_eligible_mem_bp metric requires PADDR ops */
	if (src->ops.id != DAMON_OPS_PADDR) {
		damon_for_each_scheme(scheme, src) {
			struct damos_quota *quota = &scheme->quota;

			damos_for_each_quota_goal(goal, quota) {
				if (goal->metric ==
						DAMOS_QUOTA_NODE_ELIGIBLE_MEM_BP)
					return -EINVAL;
			}
		}
	}

	err = damon_commit_schemes(dst, src);
	if (err)
		return err;
	err = damon_commit_targets(dst, src);
	if (err)
		return err;
	/*
	 * schemes and targets should be updated first, since
	 * 1. damon_set_attrs() updates monitoring results of targets and
	 * next_apply_sis of schemes, and
	 * 2. ops update should be done after pid handling is done (target
	 *    committing require putting pids).
	 */
	if (!damon_attrs_equals(&dst->attrs, &src->attrs)) {
		err = damon_set_attrs(dst, &src->attrs);
		if (err)
			return err;
	}
	dst->pause = src->pause;
	dst->ops = src->ops;
	err = damon_commit_probes(dst, src);
	if (err)
		return err;
	dst->addr_unit = src->addr_unit;
	dst->min_region_sz = src->min_region_sz;

	dst->maybe_corrupted = false;
	return 0;
}

/**
 * damon_nr_running_ctxs() - Return number of currently running contexts.
 */
int damon_nr_running_ctxs(void)
{
	int nr_ctxs;

	mutex_lock(&damon_lock);
	nr_ctxs = nr_running_ctxs;
	mutex_unlock(&damon_lock);

	return nr_ctxs;
}

/* Returns the size upper limit for each monitoring region */
static unsigned long damon_region_sz_limit(struct damon_ctx *ctx)
{
	struct damon_target *t;
	struct damon_region *r;
	unsigned long sz = 0;

	damon_for_each_target(t, ctx) {
		damon_for_each_region(r, t)
			sz += damon_sz_region(r);
	}

	if (ctx->attrs.min_nr_regions)
		sz /= ctx->attrs.min_nr_regions;
	if (sz < ctx->min_region_sz)
		sz = ctx->min_region_sz;

	return sz;
}

static void damon_split_region_at(struct damon_target *t,
				  struct damon_region *r, unsigned long sz_r);

/*
 * damon_apply_min_nr_regions() - Make effect of min_nr_regions parameter.
 * @ctx:	monitoring context.
 *
 * This function implement min_nr_regions (minimum number of damon_region
 * objects in the given monitoring context) behavior.  It first calculates
 * maximum size of each region for enforcing the min_nr_regions as total size
 * of the regions divided by the min_nr_regions.  After that, this function
 * splits regions to ensure all regions are equal to or smaller than the size
 * limit.  Finally, this function returns the maximum size limit.
 *
 * Returns: maximum size of each region for convincing min_nr_regions.
 */
static unsigned long damon_apply_min_nr_regions(struct damon_ctx *ctx)
{
	unsigned long max_region_sz = damon_region_sz_limit(ctx);
	struct damon_target *t;
	struct damon_region *r, *next;

	max_region_sz = ALIGN(max_region_sz, ctx->min_region_sz);
	damon_for_each_target(t, ctx) {
		damon_for_each_region_safe(r, next, t) {
			while (damon_sz_region(r) > max_region_sz) {
				damon_split_region_at(t, r, max_region_sz);
				r = damon_next_region(r);
			}
		}
	}
	return max_region_sz;
}

static int kdamond_fn(void *data);

/*
 * __damon_start() - Starts monitoring with given context.
 * @ctx:	monitoring context
 *
 * This function should be called while damon_lock is hold.
 *
 * Return: 0 on success, negative error code otherwise.
 */
static int __damon_start(struct damon_ctx *ctx)
{
	int err = -EBUSY;

	mutex_lock(&ctx->kdamond_lock);
	if (!ctx->kdamond) {
		err = 0;
		reinit_completion(&ctx->kdamond_started);
		ctx->kdamond = kthread_run(kdamond_fn, ctx, "kdamond.%d",
				nr_running_ctxs);
		if (IS_ERR(ctx->kdamond)) {
			err = PTR_ERR(ctx->kdamond);
			ctx->kdamond = NULL;
		} else {
			wait_for_completion(&ctx->kdamond_started);
		}
	}
	mutex_unlock(&ctx->kdamond_lock);

	return err;
}

/**
 * damon_start() - Starts the monitorings for a given group of contexts.
 * @ctxs:	an array of the pointers for contexts to start monitoring
 * @nr_ctxs:	size of @ctxs
 * @exclusive:	exclusiveness of this contexts group
 *
 * This function starts a group of monitoring threads for a group of monitoring
 * contexts.  One thread per each context is created and run in parallel.  The
 * caller should handle synchronization between the threads by itself.  If
 * @exclusive is true and a group of threads that created by other
 * 'damon_start()' call is currently running, this function does nothing but
 * returns -EBUSY.
 *
 * Return: 0 on success, negative error code otherwise.
 */
int damon_start(struct damon_ctx **ctxs, int nr_ctxs, bool exclusive)
{
	int i;
	int err = 0;

	for (i = 0; i < nr_ctxs; i++) {
		if (!is_power_of_2(ctxs[i]->min_region_sz))
			return -EINVAL;
	}

	mutex_lock(&damon_lock);
	if ((exclusive && nr_running_ctxs) ||
			(!exclusive && running_exclusive_ctxs)) {
		mutex_unlock(&damon_lock);
		return -EBUSY;
	}

	for (i = 0; i < nr_ctxs; i++) {
		err = __damon_start(ctxs[i]);
		if (err)
			break;
		nr_running_ctxs++;
	}
	if (exclusive && nr_running_ctxs)
		running_exclusive_ctxs = true;
	mutex_unlock(&damon_lock);

	return err;
}

/*
 * __damon_stop() - Stops monitoring of a given context.
 * @ctx:	monitoring context
 *
 * Return: 0 on success, negative error code otherwise.
 */
static int __damon_stop(struct damon_ctx *ctx)
{
	struct task_struct *tsk;

	mutex_lock(&ctx->kdamond_lock);
	tsk = ctx->kdamond;
	if (tsk) {
		get_task_struct(tsk);
		mutex_unlock(&ctx->kdamond_lock);
		kthread_stop_put(tsk);
		return 0;
	}
	mutex_unlock(&ctx->kdamond_lock);

	return -EPERM;
}

/**
 * damon_stop() - Stops the monitorings for a given group of contexts.
 * @ctxs:	an array of the pointers for contexts to stop monitoring
 * @nr_ctxs:	size of @ctxs
 *
 * Return: 0 on success, negative error code otherwise.
 */
int damon_stop(struct damon_ctx **ctxs, int nr_ctxs)
{
	int i, err = 0;

	for (i = 0; i < nr_ctxs; i++) {
		/* nr_running_ctxs is decremented in kdamond_fn */
		err = __damon_stop(ctxs[i]);
		if (err)
			break;
	}
	return err;
}

/**
 * damon_is_running() - Returns if a given DAMON context is running.
 * @ctx:	The DAMON context to see if running.
 *
 * Return: true if @ctx is running, false otherwise.
 */
bool damon_is_running(struct damon_ctx *ctx)
{
	bool running;

	mutex_lock(&ctx->kdamond_lock);
	running = ctx->kdamond != NULL;
	mutex_unlock(&ctx->kdamond_lock);
	return running;
}

/**
 * damon_kdamond_pid() - Return pid of a given DAMON context's worker thread.
 * @ctx:	The DAMON context of the question.
 *
 * Return: pid if @ctx is running, negative error code otherwise.
 */
int damon_kdamond_pid(struct damon_ctx *ctx)
{
	int pid = -EINVAL;

	mutex_lock(&ctx->kdamond_lock);
	if (ctx->kdamond)
		pid = ctx->kdamond->pid;
	mutex_unlock(&ctx->kdamond_lock);
	return pid;
}

/**
 * damon_call() - Invoke a given function on DAMON worker thread (kdamond).
 * @ctx:	DAMON context to call the function for.
 * @control:	Control variable of the call request.
 *
 * Ask DAMON worker thread (kdamond) of @ctx to call a function with an
 * argument data that respectively passed via &damon_call_control->fn and
 * &damon_call_control->data of @control.  If &damon_call_control->repeat of
 * @control is unset, further wait until the kdamond finishes handling of the
 * request.  Otherwise, return as soon as the request is made.
 *
 * The kdamond executes the function with the argument in the main loop, just
 * after a sampling of the iteration is finished.  The function can hence
 * safely access the internal data of the &struct damon_ctx without additional
 * synchronization.  The return value of the function will be saved in
 * &damon_call_control->return_code.
 *
 * Note that this function should be called only after damon_start() with the
 * @ctx has succeeded.  Otherwise, this function could fall into an indefinite
 * wait.
 *
 * Return: 0 on success, negative error code otherwise.
 */
int damon_call(struct damon_ctx *ctx, struct damon_call_control *control)
{
	if (!control->repeat)
		init_completion(&control->completion);
	control->canceled = false;
	INIT_LIST_HEAD(&control->list);

	mutex_lock(&ctx->call_controls_lock);
	if (ctx->call_controls_obsolete) {
		mutex_unlock(&ctx->call_controls_lock);
		return -ECANCELED;
	}
	list_add_tail(&control->list, &ctx->call_controls);
	mutex_unlock(&ctx->call_controls_lock);
	if (control->repeat)
		return 0;
	wait_for_completion(&control->completion);
	if (control->canceled)
		return -ECANCELED;
	return 0;
}

/**
 * damos_walk() - Invoke a given functions while DAMOS walk regions.
 * @ctx:	DAMON context to call the functions for.
 * @control:	Control variable of the walk request.
 *
 * Ask DAMON worker thread (kdamond) of @ctx to call a function for each region
 * that the kdamond will apply DAMOS action to, and wait until the kdamond
 * finishes handling of the request.
 *
 * The kdamond executes the given function in the main loop, for each region
 * just after it applied any DAMOS actions of @ctx to it.  The invocation is
 * made only within one &damos->apply_interval_us since damos_walk()
 * invocation, for each scheme.  The given callback function can hence safely
 * access the internal data of &struct damon_ctx and &struct damon_region that
 * each of the scheme will apply the action for next interval, without
 * additional synchronizations against the kdamond.  If every scheme of @ctx
 * passed at least one &damos->apply_interval_us, kdamond marks the request as
 * completed so that damos_walk() can wakeup and return.
 *
 * Note that this function should be called only after damon_start() with the
 * @ctx has succeeded.  Otherwise, this function could fall into an indefinite
 * wait.
 *
 * Return: 0 on success, negative error code otherwise.
 */
int damos_walk(struct damon_ctx *ctx, struct damos_walk_control *control)
{
	init_completion(&control->completion);
	control->canceled = false;
	mutex_lock(&ctx->walk_control_lock);
	if (ctx->walk_control_obsolete) {
		mutex_unlock(&ctx->walk_control_lock);
		return -ECANCELED;
	}
	if (ctx->walk_control) {
		mutex_unlock(&ctx->walk_control_lock);
		return -EBUSY;
	}
	ctx->walk_control = control;
	mutex_unlock(&ctx->walk_control_lock);
	wait_for_completion(&control->completion);
	if (control->canceled)
		return -ECANCELED;
	return 0;
}

/*
 * Warn and fix corrupted ->nr_accesses[_bp] for investigations and preventing
 * the problem being propagated.
 */
static void damon_warn_fix_nr_accesses_corruption(struct damon_region *r)
{
	if (r->nr_accesses_bp == r->nr_accesses * 10000)
		return;
	WARN_ONCE(true, "invalid nr_accesses_bp at reset: %u %u\n",
			r->nr_accesses_bp, r->nr_accesses);
	r->nr_accesses_bp = r->nr_accesses * 10000;
}

#ifdef CONFIG_DAMON_DEBUG_SANITY
static void damon_verify_reset_aggregated(struct damon_region *r,
		struct damon_ctx *c)
{
	WARN_ONCE(r->nr_accesses_bp != r->last_nr_accesses * 10000,
			"nr_accesses_bp %u last_nr_accesses %u sis %lu %lu\n",
			r->nr_accesses_bp, r->last_nr_accesses,
			c->passed_sample_intervals, c->next_aggregation_sis);
}
#else
static void damon_verify_reset_aggregated(struct damon_region *r,
		struct damon_ctx *c)
{
}
#endif


/*
 * Reset the aggregated monitoring results ('nr_accesses' of each region).
 */
static void kdamond_reset_aggregated(struct damon_ctx *c)
{
	struct damon_target *t;
	unsigned int ti = 0;	/* target's index */
	unsigned int nr_probes = 0;
	struct damon_probe *probe;

	if (trace_damon_region_aggregated_enabled()) {
		damon_for_each_probe(probe, c)
			nr_probes++;
	}

	damon_for_each_target(t, c) {
		struct damon_region *r;

		damon_for_each_region(r, t) {
			int i;

			trace_damon_aggregated(ti, r, damon_nr_regions(t));
			trace_damon_region_aggregated(ti, r,
					damon_nr_regions(t), nr_probes);
			damon_warn_fix_nr_accesses_corruption(r);
			r->last_nr_accesses = r->nr_accesses;
			r->nr_accesses = 0;
			for (i = 0; i < DAMON_MAX_PROBES; i++)
				r->probe_hits[i] = 0;
			damon_verify_reset_aggregated(r, c);
		}
		ti++;
	}
}

static unsigned long damon_get_intervals_score(struct damon_ctx *c)
{
	struct damon_target *t;
	struct damon_region *r;
	unsigned long sz_region, max_access_events = 0, access_events = 0;
	unsigned long target_access_events;
	unsigned long goal_bp = c->attrs.intervals_goal.access_bp;

	damon_for_each_target(t, c) {
		damon_for_each_region(r, t) {
			sz_region = damon_sz_region(r);
			max_access_events += sz_region * c->attrs.aggr_samples;
			access_events += sz_region * r->nr_accesses;
		}
	}
	target_access_events = max_access_events * goal_bp / 10000;
	target_access_events = target_access_events ? : 1;
	return mult_frac(access_events, 10000, target_access_events);
}

static unsigned long damon_feed_loop_next_input(unsigned long last_input,
		unsigned long score);

static unsigned long damon_get_intervals_adaptation_bp(struct damon_ctx *c)
{
	unsigned long score_bp, adaptation_bp;

	score_bp = damon_get_intervals_score(c);
	adaptation_bp = damon_feed_loop_next_input(100000000, score_bp) /
		10000;
	/*
	 * adaptation_bp ranges from 1 to 20,000.  Avoid too rapid reduction of
	 * the intervals by rescaling [1,10,000] to [5000, 10,000].
	 */
	if (adaptation_bp <= 10000)
		adaptation_bp = 5000 + adaptation_bp / 2;
	return adaptation_bp;
}

static void kdamond_tune_intervals(struct damon_ctx *c)
{
	unsigned long adaptation_bp;
	struct damon_attrs new_attrs;
	struct damon_intervals_goal *goal;

	adaptation_bp = damon_get_intervals_adaptation_bp(c);
	if (adaptation_bp == 10000)
		return;

	new_attrs = c->attrs;
	goal = &c->attrs.intervals_goal;
	new_attrs.sample_interval = min(goal->max_sample_us,
			c->attrs.sample_interval * adaptation_bp / 10000);
	new_attrs.sample_interval = max(goal->min_sample_us,
			new_attrs.sample_interval);
	new_attrs.aggr_interval = new_attrs.sample_interval *
		c->attrs.aggr_samples;
	trace_damon_monitor_intervals_tune(new_attrs.sample_interval);
	damon_set_attrs(c, &new_attrs);
}

static bool __damos_valid_target(struct damon_region *r, struct damos *s)
{
	unsigned long sz;
	unsigned int nr_accesses = r->nr_accesses_bp / 10000;

	sz = damon_sz_region(r);
	return s->pattern.min_sz_region <= sz &&
		sz <= s->pattern.max_sz_region &&
		s->pattern.min_nr_accesses <= nr_accesses &&
		nr_accesses <= s->pattern.max_nr_accesses &&
		s->pattern.min_age_region <= r->age &&
		r->age <= s->pattern.max_age_region;
}

/*
 * damos_quota_is_set() - Return if the given quota is actually set.
 * @quota:	The quota to check.
 *
 * Returns true if the quota is set, false otherwise.
 */
static bool damos_quota_is_set(struct damos_quota *quota)
{
	return quota->esz || quota->sz || quota->ms ||
		!damos_quota_goals_empty(quota);
}

static bool damos_valid_target(struct damon_ctx *c, struct damon_region *r,
		struct damos *s)
{
	bool ret = __damos_valid_target(r, s);

	if (!ret || !damos_quota_is_set(&s->quota) || !c->ops.get_scheme_score)
		return ret;

	return c->ops.get_scheme_score(c, r, s) >= s->quota.min_score;
}

/*
 * damos_skip_charged_region() - Check if the given region or starting part of
 * it is already charged for the DAMOS quota.
 * @t:	The target of the region.
 * @rp:	The pointer to the region.
 * @s:	The scheme to be applied.
 * @min_region_sz:	minimum region size.
 *
 * If a quota of a scheme has exceeded in a quota charge window, the scheme's
 * action would applied to only a part of the target access pattern fulfilling
 * regions.  To avoid applying the scheme action to only already applied
 * regions, DAMON skips applying the scheme action to the regions that charged
 * in the previous charge window.
 *
 * This function checks if a given region should be skipped or not for the
 * reason.  If only the starting part of the region has previously charged,
 * this function splits the region into two so that the second one covers the
 * area that not charged in the previous charge widnow, and return true.  The
 * caller can see the second one on the next iteration of the region walk.
 * Note that this means the caller should use damon_for_each_region() instead
 * of damon_for_each_region_safe().  If damon_for_each_region_safe() is used,
 * the second region will just be ignored.
 *
 * Return: true if the region should be skipped, false otherwise.
 */
static bool damos_skip_charged_region(struct damon_target *t,
		struct damon_region *r, struct damos *s,
		unsigned long min_region_sz)
{
	struct damos_quota *quota = &s->quota;
	unsigned long sz_to_skip;

	/* Skip previously charged regions */
	if (quota->charge_target_from) {
		if (t != quota->charge_target_from)
			return true;
		if (r == damon_last_region(t)) {
			quota->charge_target_from = NULL;
			quota->charge_addr_from = 0;
			return true;
		}
		if (quota->charge_addr_from &&
				r->ar.end <= quota->charge_addr_from)
			return true;

		if (quota->charge_addr_from && r->ar.start <
				quota->charge_addr_from) {
			sz_to_skip = ALIGN_DOWN(quota->charge_addr_from -
					r->ar.start, min_region_sz);
			if (!sz_to_skip) {
				if (damon_sz_region(r) <= min_region_sz)
					return true;
				sz_to_skip = min_region_sz;
			}
			damon_split_region_at(t, r, sz_to_skip);
			return true;
		}
		quota->charge_target_from = NULL;
		quota->charge_addr_from = 0;
	}
	return false;
}

static void damos_update_stat(struct damos *s,
		unsigned long sz_tried, unsigned long sz_applied,
		unsigned long sz_ops_filter_passed)
{
	s->stat.nr_tried++;
	s->stat.sz_tried += sz_tried;
	if (sz_applied)
		s->stat.nr_applied++;
	s->stat.sz_applied += sz_applied;
	s->stat.sz_ops_filter_passed += sz_ops_filter_passed;
}

static bool damos_filter_match(struct damon_ctx *ctx, struct damon_target *t,
		struct damon_region *r, struct damos_filter *filter,
		unsigned long min_region_sz)
{
	bool matched = false;
	struct damon_target *ti;
	int target_idx = 0;
	unsigned long start, end;

	switch (filter->type) {
	case DAMOS_FILTER_TYPE_TARGET:
		damon_for_each_target(ti, ctx) {
			if (ti == t)
				break;
			target_idx++;
		}
		matched = target_idx == filter->target_idx;
		break;
	case DAMOS_FILTER_TYPE_ADDR:
		start = ALIGN_DOWN(filter->addr_range.start, min_region_sz);
		end = ALIGN_DOWN(filter->addr_range.end, min_region_sz);

		/* inside the range */
		if (start <= r->ar.start && r->ar.end <= end) {
			matched = true;
			break;
		}
		/* outside of the range */
		if (r->ar.end <= start || end <= r->ar.start) {
			matched = false;
			break;
		}
		/* start before the range and overlap */
		if (r->ar.start < start) {
			damon_split_region_at(t, r, start - r->ar.start);
			matched = false;
			break;
		}
		/* start inside the range */
		damon_split_region_at(t, r, end - r->ar.start);
		matched = true;
		break;
	default:
		return false;
	}

	return matched == filter->matching;
}

static bool damos_core_filter_out(struct damon_ctx *ctx, struct damon_target *t,
		struct damon_region *r, struct damos *s)
{
	struct damos_filter *filter;

	s->core_filters_allowed = false;
	damos_for_each_core_filter(filter, s) {
		if (damos_filter_match(ctx, t, r, filter, ctx->min_region_sz)) {
			if (filter->allow)
				s->core_filters_allowed = true;
			return !filter->allow;
		}
	}
	return s->core_filters_default_reject;
}

/*
 * damos_walk_call_walk() - Call &damos_walk_control->walk_fn.
 * @ctx:	The context of &damon_ctx->walk_control.
 * @t:		The monitoring target of @r that @s will be applied.
 * @r:		The region of @t that @s will be applied.
 * @s:		The scheme of @ctx that will be applied to @r.
 *
 * This function is called from kdamond whenever it asked the operation set to
 * apply a DAMOS scheme action to a region.  If a DAMOS walk request is
 * installed by damos_walk() and not yet uninstalled, invoke it.
 */
static void damos_walk_call_walk(struct damon_ctx *ctx, struct damon_target *t,
		struct damon_region *r, struct damos *s,
		unsigned long sz_filter_passed)
{
	struct damos_walk_control *control;

	if (s->walk_completed)
		return;

	control = ctx->walk_control;
	if (!control)
		return;

	control->walk_fn(control->data, ctx, t, r, s, sz_filter_passed);
}

/*
 * damos_walk_complete() - Complete DAMOS walk request if all walks are done.
 * @ctx:	The context of &damon_ctx->walk_control.
 * @s:		A scheme of @ctx that all walks are now done.
 *
 * This function is called when kdamond finished applying the action of a DAMOS
 * scheme to all regions that eligible for the given &damos->apply_interval_us.
 * If every scheme of @ctx including @s now finished walking for at least one
 * &damos->apply_interval_us, this function makrs the handling of the given
 * DAMOS walk request is done, so that damos_walk() can wake up and return.
 */
static void damos_walk_complete(struct damon_ctx *ctx, struct damos *s)
{
	struct damos *siter;
	struct damos_walk_control *control;

	control = ctx->walk_control;
	if (!control)
		return;

	s->walk_completed = true;
	/* if all schemes completed, signal completion to walker */
	damon_for_each_scheme(siter, ctx) {
		if (!siter->walk_completed)
			return;
	}
	damon_for_each_scheme(siter, ctx)
		siter->walk_completed = false;

	complete(&control->completion);
	ctx->walk_control = NULL;
}

/*
 * damos_walk_cancel() - Cancel the current DAMOS walk request.
 * @ctx:	The context of &damon_ctx->walk_control.
 *
 * This function is called when @ctx is deactivated by DAMOS watermarks, DAMOS
 * walk is requested but there is no DAMOS scheme to walk for, or the kdamond
 * is already out of the main loop and therefore gonna be terminated, and hence
 * cannot continue the walks.  This function therefore marks the walk request
 * as canceled, so that damos_walk() can wake up and return.
 */
static void damos_walk_cancel(struct damon_ctx *ctx)
{
	struct damos_walk_control *control;

	mutex_lock(&ctx->walk_control_lock);
	control = ctx->walk_control;
	mutex_unlock(&ctx->walk_control_lock);

	if (!control)
		return;
	control->canceled = true;
	complete(&control->completion);
	mutex_lock(&ctx->walk_control_lock);
	ctx->walk_control = NULL;
	mutex_unlock(&ctx->walk_control_lock);
}

static void damos_charge_quota(struct damos_quota *quota,
		unsigned long sz_region, unsigned long sz_applied)
{
	/*
	 * sz_applied could be bigger than sz_region, depending on ops
	 * implementation of the action, e.g., damos_pa_pageout().  Charge only
	 * the region size in the case.
	 */
	if (!quota->fail_charge_denom || sz_applied > sz_region)
		quota->charged_sz += sz_region;
	else
		quota->charged_sz += sz_applied + mult_frac(
				(sz_region - sz_applied),
				quota->fail_charge_num,
				quota->fail_charge_denom);
}

static bool damos_quota_is_full(struct damos_quota *quota,
		unsigned long min_region_sz)
{
	if (!damos_quota_is_set(quota))
		return false;
	if (quota->charged_sz >= quota->esz)
		return true;
	/*
	 * DAMOS action is applied per region, so <min_region_sz remaining
	 * quota means the quota is effectively full.
	 */
	return quota->esz - quota->charged_sz < min_region_sz;
}

static void damos_apply_scheme(struct damon_ctx *c, struct damon_target *t,
		struct damon_region *r, struct damos *s)
{
	struct damos_quota *quota = &s->quota;
	unsigned long sz = damon_sz_region(r);
	struct timespec64 begin, end;
	unsigned long sz_applied = 0;
	unsigned long sz_ops_filter_passed = 0;
	/*
	 * We plan to support multiple context per kdamond, as DAMON sysfs
	 * implies with 'nr_contexts' file.  Nevertheless, only single context
	 * per kdamond is supported for now.  So, we can simply use '0' context
	 * index here.
	 */
	unsigned int cidx = 0;
	struct damos *siter;		/* schemes iterator */
	unsigned int sidx = 0;
	struct damon_target *titer;	/* targets iterator */
	unsigned int tidx = 0;
	bool do_trace = false;

	/* get indices for trace_damos_before_apply() */
	if (trace_damos_before_apply_enabled()) {
		damon_for_each_scheme(siter, c) {
			if (siter == s)
				break;
			sidx++;
		}
		damon_for_each_target(titer, c) {
			if (titer == t)
				break;
			tidx++;
		}
		do_trace = true;
	}

	if (c->ops.apply_scheme) {
		if (damos_quota_is_set(quota) &&
				quota->charged_sz + sz > quota->esz) {
			sz = ALIGN_DOWN(quota->esz - quota->charged_sz,
					c->min_region_sz);
			if (!sz)
				goto update_stat;
			damon_split_region_at(t, r, sz);
		}
		if (damos_core_filter_out(c, t, r, s))
			return;
		ktime_get_coarse_ts64(&begin);
		trace_damos_before_apply(cidx, sidx, tidx, r,
				damon_nr_regions(t), do_trace);
		sz_applied = c->ops.apply_scheme(c, t, r, s,
				&sz_ops_filter_passed);
		damos_walk_call_walk(c, t, r, s, sz_ops_filter_passed);
		ktime_get_coarse_ts64(&end);
		quota->total_charged_ns += timespec64_to_ns(&end) -
			timespec64_to_ns(&begin);
		damos_charge_quota(quota, sz, sz_applied);
		if (damos_quota_is_full(quota, c->min_region_sz)) {
			quota->charge_target_from = t;
			quota->charge_addr_from = r->ar.end;
		}
	}
	if (s->action != DAMOS_STAT)
		r->age = 0;

update_stat:
	damos_update_stat(s, sz, sz_applied, sz_ops_filter_passed);
}

static void damon_do_apply_schemes(struct damon_ctx *c,
				   struct damon_target *t,
				   struct damon_region *r)
{
	struct damos *s;

	damon_for_each_scheme(s, c) {
		struct damos_quota *quota = &s->quota;

		if (time_before(c->passed_sample_intervals, s->next_apply_sis))
			continue;

		if (!s->wmarks.activated)
			continue;

		/* Check the quota */
		if (damos_quota_is_full(quota, c->min_region_sz))
			continue;

		if (damos_skip_charged_region(t, r, s, c->min_region_sz))
			continue;

		if (s->max_nr_snapshots &&
				s->max_nr_snapshots <= s->stat.nr_snapshots)
			continue;

		if (damos_valid_target(c, r, s))
			damos_apply_scheme(c, t, r, s);

		if (damon_is_last_region(r, t))
			s->stat.nr_snapshots++;
	}
}

/*
 * damos_apply_target() - Apply DAMOS schemes to a given target.
 * @c:			monitoring context to apply its DAMOS schemes to..
 * @t:			monitoring target to apply the schemes to.
 * @max_region_sz:	maximum region size for @c.
 *
 * This function could split regions for keeping the quota.  To minimize
 * overhead from the split operations increased number of regions, this
 * function will also merge regions after the schemes applying attempt is done,
 * for each region.  The merge operation is made only when it doesn't lose the
 * monitoring information and not violating @max_region_sz.
 *
 * Hence, after this function is called, the total number of regions could
 * be increased or reduced.  The increase could make max_nr_regions temporarily
 * be violated, until the next per-aggregation interval regions merge operation
 * is executed.  The decrease will not violate min_nr_regions though, since it
 * keeps @max_region_sz.
 */
static void damos_apply_target(struct damon_ctx *c, struct damon_target *t,
		unsigned long max_region_sz)
{
	struct damon_region *r;

	damon_for_each_region(r, t) {
		struct damon_region *prev_r;

		damon_do_apply_schemes(c, t, r);
		/*
		 * damon_do_apply_scheems() could split the region for the
		 * quota.  Keeping the new slices is an overhead.  Merge back
		 * the slices into the previous region if it doesn't lose any
		 * information and not violating the max_region_sz.
		 */
		if (damon_first_region(t) == r)
			continue;
		prev_r = damon_prev_region(r);
		if (prev_r->ar.end != r->ar.start)
			continue;
		if (prev_r->age != r->age)
			continue;
		if (prev_r->last_nr_accesses != r->last_nr_accesses)
			continue;
		if (prev_r->nr_accesses != r->nr_accesses)
			continue;
		if (r->ar.end - prev_r->ar.start > max_region_sz)
			continue;
		prev_r->ar.end = r->ar.end;
		damon_destroy_region(r, t);
		r = prev_r;
	}
}

/*
 * damon_feed_loop_next_input() - get next input to achieve a target score.
 * @last_input	The last input.
 * @score	Current score that made with @last_input.
 *
 * Calculate next input to achieve the target score, based on the last input
 * and current score.  Assuming the input and the score are positively
 * proportional, calculate how much compensation should be added to or
 * subtracted from the last input as a proportion of the last input.  Avoid
 * next input always being zero by setting it non-zero always.  In short form
 * (assuming support of float and signed calculations), the algorithm is as
 * below.
 *
 * next_input = max(last_input * ((goal - current) / goal + 1), 1)
 *
 * For simple implementation, we assume the target score is always 10,000.  The
 * caller should adjust @score for this.
 *
 * Returns next input that assumed to achieve the target score.
 */
static unsigned long damon_feed_loop_next_input(unsigned long last_input,
		unsigned long score)
{
	const unsigned long goal = 10000;
	/* Set minimum input as 10000 to avoid compensation be zero */
	const unsigned long min_input = 10000;
	unsigned long score_goal_diff, compensation;
	bool over_achieving = score > goal;

	if (score == goal)
		return last_input;
	if (score >= goal * 2)
		return min_input;

	if (over_achieving)
		score_goal_diff = score - goal;
	else
		score_goal_diff = goal - score;

	if (last_input < ULONG_MAX / score_goal_diff)
		compensation = last_input * score_goal_diff / goal;
	else
		compensation = last_input / goal * score_goal_diff;

	if (over_achieving)
		return max(last_input - compensation, min_input);
	if (last_input < ULONG_MAX - compensation)
		return last_input + compensation;
	return ULONG_MAX;
}

#ifdef CONFIG_PSI

static u64 damos_get_some_mem_psi_total(void)
{
	if (static_branch_likely(&psi_disabled))
		return 0;
	return div_u64(psi_system.total[PSI_AVGS][PSI_MEM * 2],
			NSEC_PER_USEC);
}

#else	/* CONFIG_PSI */

static inline u64 damos_get_some_mem_psi_total(void)
{
	return 0;
};

#endif	/* CONFIG_PSI */

#ifdef CONFIG_NUMA
static bool invalid_mem_node(int nid)
{
	return nid < 0 || nid >= MAX_NUMNODES || !node_state(nid, N_MEMORY);
}

static __kernel_ulong_t damos_get_node_mem_bp(
		struct damos_quota_goal *goal)
{
	struct sysinfo i;
	__kernel_ulong_t numerator;

	if (invalid_mem_node(goal->nid)) {
		if (goal->metric == DAMOS_QUOTA_NODE_MEM_USED_BP)
			return 0;
		else	/* DAMOS_QUOTA_NODE_MEM_FREE_BP */
			return 10000;
	}

	si_meminfo_node(&i, goal->nid);
	if (goal->metric == DAMOS_QUOTA_NODE_MEM_USED_BP)
		numerator = i.totalram - i.freeram;
	else	/* DAMOS_QUOTA_NODE_MEM_FREE_BP */
		numerator = i.freeram;
	return mult_frac(numerator, 10000, i.totalram);
}

static unsigned long damos_get_node_memcg_used_bp(
		struct damos_quota_goal *goal)
{
	struct mem_cgroup *memcg;
	struct lruvec *lruvec;
	unsigned long used_pages, numerator;
	struct sysinfo i;

	if (invalid_mem_node(goal->nid)) {
		if (goal->metric == DAMOS_QUOTA_NODE_MEMCG_USED_BP)
			return 0;
		else	/* DAMOS_QUOTA_NODE_MEMCG_FREE_BP */
			return 10000;
	}

	memcg = mem_cgroup_get_from_id(goal->memcg_id);
	if (!memcg) {
		if (goal->metric == DAMOS_QUOTA_NODE_MEMCG_USED_BP)
			return 0;
		else	/* DAMOS_QUOTA_NODE_MEMCG_FREE_BP */
			return 10000;
	}

	mem_cgroup_flush_stats(memcg);
	lruvec = mem_cgroup_lruvec(memcg, NODE_DATA(goal->nid));
	used_pages = lruvec_page_state(lruvec, NR_ACTIVE_ANON);
	used_pages += lruvec_page_state(lruvec, NR_INACTIVE_ANON);
	used_pages += lruvec_page_state(lruvec, NR_ACTIVE_FILE);
	used_pages += lruvec_page_state(lruvec, NR_INACTIVE_FILE);

	mem_cgroup_put(memcg);

	si_meminfo_node(&i, goal->nid);
	if (goal->metric == DAMOS_QUOTA_NODE_MEMCG_USED_BP)
		numerator = used_pages;
	else	/* DAMOS_QUOTA_NODE_MEMCG_FREE_BP */
		numerator = i.totalram - used_pages;
	return mult_frac(numerator, 10000, i.totalram);
}

#ifdef CONFIG_DAMON_PADDR
/*
 * damos_calc_eligible_bytes() - Calculate raw eligible bytes per node.
 * @c:		The DAMON context.
 * @s:		The scheme.
 * @nid:	The target NUMA node id.
 * @total:	Output for total eligible bytes across all nodes.
 *
 * Iterates through each folio in eligible regions to accurately determine
 * which node the memory resides on. Returns eligible bytes on the specified
 * node and sets *total to the sum across all nodes.
 *
 * Note: This function requires damon_get_folio() from ops-common.c, which is
 * only available when CONFIG_DAMON_PADDR is enabled. It also requires the
 * context to be using PADDR operations for meaningful results.
 */
static phys_addr_t damos_calc_eligible_bytes(struct damon_ctx *c,
		struct damos *s, int nid, phys_addr_t *total)
{
	struct damon_target *t;
	struct damon_region *r;
	phys_addr_t total_eligible = 0;
	phys_addr_t node_eligible = 0;

	damon_for_each_target(t, c) {
		damon_for_each_region(r, t) {
			phys_addr_t addr, end_addr;

			if (!__damos_valid_target(r, s))
				continue;

			/* Convert from core address units to physical bytes */
			addr = (phys_addr_t)r->ar.start * c->addr_unit;
			end_addr = (phys_addr_t)r->ar.end * c->addr_unit;
			while (addr < end_addr) {
				struct folio *folio;
				phys_addr_t folio_start, folio_end;
				phys_addr_t overlap_start, overlap_end;
				phys_addr_t counted;

				folio = damon_get_folio(PHYS_PFN(addr));
				if (!folio) {
					addr = PAGE_ALIGN_DOWN(addr +
							PAGE_SIZE);
					if (!addr)
						break;
					continue;
				}

				/*
				 * Calculate exact overlap between the region
				 * [addr, end_addr) and the folio range.
				 * The folio may start before addr if addr is
				 * in the middle of a large folio.
				 */
				folio_start = PFN_PHYS(folio_pfn(folio));
				folio_end = folio_start + folio_size(folio);

				overlap_start = max(addr, folio_start);
				overlap_end = min(end_addr, folio_end);

				if (overlap_end > overlap_start) {
					counted = overlap_end - overlap_start;
					total_eligible += counted;
					if (folio_nid(folio) == nid)
						node_eligible += counted;
				}

				/* Advance past the entire folio */
				addr = folio_end;
				folio_put(folio);
			}
			cond_resched();
		}
	}

	*total = total_eligible;
	return node_eligible;
}

static unsigned long damos_get_node_eligible_mem_bp(struct damon_ctx *c,
		struct damos *s, int nid)
{
	phys_addr_t total_eligible = 0;
	phys_addr_t node_eligible;

	if (c->ops.id != DAMON_OPS_PADDR)
		return 0;

	if (nid < 0 || nid >= MAX_NUMNODES || !node_online(nid))
		return 0;

	node_eligible = damos_calc_eligible_bytes(c, s, nid, &total_eligible);

	if (!(unsigned long)total_eligible)
		return 0;

	return mult_frac((unsigned long)node_eligible, 10000,
			(unsigned long)total_eligible);
}
#else /* CONFIG_DAMON_PADDR */
static unsigned long damos_get_node_eligible_mem_bp(struct damon_ctx *c,
		struct damos *s, int nid)
{
	return 0;
}
#endif /* CONFIG_DAMON_PADDR */
#else /* CONFIG_NUMA */
static __kernel_ulong_t damos_get_node_mem_bp(
		struct damos_quota_goal *goal)
{
	return 0;
}

static unsigned long damos_get_node_memcg_used_bp(
		struct damos_quota_goal *goal)
{
	return 0;
}

static unsigned long damos_get_node_eligible_mem_bp(struct damon_ctx *c,
		struct damos *s, int nid)
{
	return 0;
}
#endif /* CONFIG_NUMA */

/*
 * Returns LRU-active or inactive memory to total LRU memory size ratio.
 */
static unsigned int damos_get_in_active_mem_bp(bool active_ratio)
{
	unsigned long active, inactive, total;

	/* This should align with /proc/meminfo output */
	active = global_node_page_state(NR_LRU_BASE + LRU_ACTIVE_ANON) +
		global_node_page_state(NR_LRU_BASE + LRU_ACTIVE_FILE);
	inactive = global_node_page_state(NR_LRU_BASE + LRU_INACTIVE_ANON) +
		global_node_page_state(NR_LRU_BASE + LRU_INACTIVE_FILE);
	total = active + inactive;
	if (active_ratio)
		return mult_frac(active, 10000, total);
	return mult_frac(inactive, 10000, total);
}

static void damos_set_quota_goal_current_value(struct damon_ctx *c,
		struct damos *s, struct damos_quota_goal *goal)
{
	u64 now_psi_total;

	switch (goal->metric) {
	case DAMOS_QUOTA_USER_INPUT:
		/* User should already set goal->current_value */
		break;
	case DAMOS_QUOTA_SOME_MEM_PSI_US:
		now_psi_total = damos_get_some_mem_psi_total();
		goal->current_value = now_psi_total - goal->last_psi_total;
		goal->last_psi_total = now_psi_total;
		break;
	case DAMOS_QUOTA_NODE_MEM_USED_BP:
	case DAMOS_QUOTA_NODE_MEM_FREE_BP:
		goal->current_value = damos_get_node_mem_bp(goal);
		break;
	case DAMOS_QUOTA_NODE_MEMCG_USED_BP:
	case DAMOS_QUOTA_NODE_MEMCG_FREE_BP:
		goal->current_value = damos_get_node_memcg_used_bp(goal);
		break;
	case DAMOS_QUOTA_ACTIVE_MEM_BP:
	case DAMOS_QUOTA_INACTIVE_MEM_BP:
		goal->current_value = damos_get_in_active_mem_bp(
				goal->metric == DAMOS_QUOTA_ACTIVE_MEM_BP);
		break;
	case DAMOS_QUOTA_NODE_ELIGIBLE_MEM_BP:
		goal->current_value = damos_get_node_eligible_mem_bp(c, s,
				goal->nid);
		break;
	default:
		break;
	}
}

/* Return the highest score since it makes schemes least aggressive */
static unsigned long damos_quota_score(struct damon_ctx *c, struct damos *s)
{
	struct damos_quota_goal *goal;
	struct damos_quota *quota = &s->quota;
	unsigned long highest_score = 0;

	damos_for_each_quota_goal(goal, quota) {
		damos_set_quota_goal_current_value(c, s, goal);
		highest_score = max(highest_score,
				mult_frac(goal->current_value, 10000,
					goal->target_value));
	}

	return highest_score;
}

static void damos_goal_tune_esz_bp_consist(struct damon_ctx *c, struct damos *s)
{
	struct damos_quota *quota = &s->quota;
	unsigned long score = damos_quota_score(c, s);

	quota->esz_bp = damon_feed_loop_next_input(
			max(quota->esz_bp, 10000UL), score);
}

static void damos_goal_tune_esz_bp_temporal(struct damon_ctx *c,
		struct damos *s)
{
	struct damos_quota *quota = &s->quota;
	unsigned long score = damos_quota_score(c, s);

	if (score >= 10000)
		quota->esz_bp = 0;
	else if (quota->sz)
		quota->esz_bp = quota->sz * 10000;
	else
		quota->esz_bp = ULONG_MAX;
}

/*
 * Called only if quota->ms, or quota->sz are set, or quota->goals is not empty
 */
static void damos_set_effective_quota(struct damon_ctx *ctx, struct damos *s)
{
	struct damos_quota *quota = &s->quota;
	unsigned long throughput;
	unsigned long esz = ULONG_MAX;

	if (!quota->ms && list_empty(&quota->goals)) {
		quota->esz = quota->sz;
		return;
	}

	if (!list_empty(&quota->goals)) {
		if (quota->goal_tuner == DAMOS_QUOTA_GOAL_TUNER_CONSIST)
			damos_goal_tune_esz_bp_consist(ctx, s);
		else if (quota->goal_tuner == DAMOS_QUOTA_GOAL_TUNER_TEMPORAL)
			damos_goal_tune_esz_bp_temporal(ctx, s);
		esz = quota->esz_bp / 10000;
	}

	if (quota->ms) {
		if (quota->total_charged_ns)
			throughput = mult_frac(quota->total_charged_sz,
					1000000, quota->total_charged_ns);
		else
			throughput = PAGE_SIZE * 1024;
		esz = min(throughput * quota->ms, esz);
		esz = max(ctx->min_region_sz, esz);
	}

	if (quota->sz && quota->sz < esz)
		esz = quota->sz;

	quota->esz = esz;
}

static void damos_trace_esz(struct damon_ctx *c, struct damos *s,
		struct damos_quota *quota)
{
	unsigned int cidx = 0, sidx = 0;
	struct damos *siter;

	damon_for_each_scheme(siter, c) {
		if (siter == s)
			break;
		sidx++;
	}
	trace_damos_esz(cidx, sidx, quota->esz);
}

static void damos_adjust_quota(struct damon_ctx *c, struct damos *s)
{
	struct damos_quota *quota = &s->quota;
	struct damon_target *t;
	struct damon_region *r;
	unsigned long cumulated_sz, cached_esz;
	unsigned int score, max_score = 0;

	if (!quota->ms && !quota->sz && list_empty(&quota->goals))
		return;

	/* First charge window */
	if (!quota->total_charged_sz && !quota->charged_from) {
		quota->charged_from = jiffies;
		damos_set_effective_quota(c, s);
		if (trace_damos_esz_enabled())
			damos_trace_esz(c, s, quota);
	}

	/* New charge window starts */
	if (!time_in_range_open(jiffies, quota->charged_from,
				quota->charged_from +
				msecs_to_jiffies(quota->reset_interval))) {
		if (damos_quota_is_full(quota, c->min_region_sz))
			s->stat.qt_exceeds++;
		quota->total_charged_sz += quota->charged_sz;
		quota->charged_from = jiffies;
		quota->charged_sz = 0;
		if (trace_damos_esz_enabled())
			cached_esz = quota->esz;
		damos_set_effective_quota(c, s);
		if (trace_damos_esz_enabled() && quota->esz != cached_esz)
			damos_trace_esz(c, s, quota);
	}

	if (!c->ops.get_scheme_score)
		return;

	/* Fill up the score histogram */
	memset(c->regions_score_histogram, 0,
			sizeof(*c->regions_score_histogram) *
			(DAMOS_MAX_SCORE + 1));
	damon_for_each_target(t, c) {
		damon_for_each_region(r, t) {
			if (!__damos_valid_target(r, s))
				continue;
			if (damos_core_filter_out(c, t, r, s))
				continue;
			score = c->ops.get_scheme_score(c, r, s);
			c->regions_score_histogram[score] +=
				damon_sz_region(r);
			if (score > max_score)
				max_score = score;
		}
	}

	/* Set the min score limit */
	for (cumulated_sz = 0, score = max_score; ; score--) {
		cumulated_sz += c->regions_score_histogram[score];
		if (cumulated_sz >= quota->esz || !score)
			break;
	}
	quota->min_score = score;
}

static void damos_trace_stat(struct damon_ctx *c, struct damos *s)
{
	unsigned int cidx = 0, sidx = 0;
	struct damos *siter;

	if (!trace_damos_stat_after_apply_interval_enabled())
		return;

	damon_for_each_scheme(siter, c) {
		if (siter == s)
			break;
		sidx++;
	}
	trace_call__damos_stat_after_apply_interval(cidx, sidx, &s->stat);
}

static void kdamond_apply_schemes(struct damon_ctx *c)
{
	struct damon_target *t;
	struct damos *s;
	bool has_schemes_to_apply = false;
	unsigned long max_region_sz;

	damon_for_each_scheme(s, c) {
		if (time_before(c->passed_sample_intervals, s->next_apply_sis))
			continue;

		if (!s->wmarks.activated)
			continue;

		has_schemes_to_apply = true;

		damos_adjust_quota(c, s);
	}

	if (!has_schemes_to_apply)
		return;

	max_region_sz = damon_region_sz_limit(c);
	mutex_lock(&c->walk_control_lock);
	damon_for_each_target(t, c) {
		if (c->ops.target_valid && c->ops.target_valid(t) == false)
			continue;
		damos_apply_target(c, t, max_region_sz);
	}

	damon_for_each_scheme(s, c) {
		if (time_before(c->passed_sample_intervals, s->next_apply_sis))
			continue;
		damos_walk_complete(c, s);
		damos_set_next_apply_sis(s, c);
		s->last_applied = NULL;
		damos_trace_stat(c, s);
	}
	mutex_unlock(&c->walk_control_lock);
}

#ifdef CONFIG_DAMON_DEBUG_SANITY
static void damon_verify_merge_two_regions(
		struct damon_region *l, struct damon_region *r)
{
	/* damon_merge_two_regions() may created incorrect left region */
	WARN_ONCE(l->ar.start >= l->ar.end, "l: %lu-%lu, r: %lu-%lu\n",
			l->ar.start, l->ar.end, r->ar.start, r->ar.end);
}
#else
static void damon_verify_merge_two_regions(
		struct damon_region *l, struct damon_region *r)
{
}
#endif

/*
 * Merge two adjacent regions into one region
 */
static void damon_merge_two_regions(struct damon_target *t,
		struct damon_region *l, struct damon_region *r)
{
	unsigned long sz_l = damon_sz_region(l), sz_r = damon_sz_region(r);
	int i;

	l->nr_accesses = (l->nr_accesses * sz_l + r->nr_accesses * sz_r) /
			(sz_l + sz_r);
	l->nr_accesses_bp = l->nr_accesses * 10000;
	l->age = (l->age * sz_l + r->age * sz_r) / (sz_l + sz_r);
	l->ar.end = r->ar.end;
	/* todo: do this for only installed probes */
	for (i = 0; i < DAMON_MAX_PROBES; i++)
		l->probe_hits[i] = (l->probe_hits[i] * sz_l + r->probe_hits[i]
				* sz_r) / (sz_l + sz_r);
	damon_verify_merge_two_regions(l, r);
	damon_destroy_region(r, t);
}

#ifdef CONFIG_DAMON_DEBUG_SANITY
static void damon_verify_merge_regions_of(struct damon_region *r)
{
	WARN_ONCE(r->nr_accesses != r->nr_accesses_bp / 10000,
			"nr_accesses (%u) != nr_accesses_bp (%u)\n",
			r->nr_accesses, r->nr_accesses_bp);
}
#else
static void damon_verify_merge_regions_of(struct damon_region *r)
{
}
#endif


/*
 * Merge adjacent regions having similar access frequencies
 *
 * t		target affected by this merge operation
 * thres	'->nr_accesses' diff threshold for the merge
 * sz_limit	size upper limit of each region
 */
static void damon_merge_regions_of(struct damon_target *t, unsigned int thres,
				   unsigned long sz_limit)
{
	struct damon_region *r, *prev = NULL, *next;

	damon_for_each_region_safe(r, next, t) {
		damon_verify_merge_regions_of(r);
		if (abs(r->nr_accesses - r->last_nr_accesses) > thres)
			r->age = 0;
		else if ((r->nr_accesses == 0) != (r->last_nr_accesses == 0))
			r->age = 0;
		else
			r->age++;

		if (prev && prev->ar.end == r->ar.start &&
		    abs(prev->nr_accesses - r->nr_accesses) <= thres &&
		    damon_sz_region(prev) + damon_sz_region(r) <= sz_limit)
			damon_merge_two_regions(t, prev, r);
		else
			prev = r;
	}
}

/*
 * Merge adjacent regions having similar access frequencies
 *
 * threshold	'->nr_accesses' diff threshold for the merge
 * sz_limit	size upper limit of each region
 *
 * This function merges monitoring target regions which are adjacent and their
 * access frequencies are similar.  This is for minimizing the monitoring
 * overhead under the dynamically changeable access pattern.  If a merge was
 * unnecessarily made, later 'kdamond_split_regions()' will revert it.
 *
 * The total number of regions could be higher than the user-defined limit,
 * max_nr_regions for some cases.  For example, the user can update
 * max_nr_regions to a number that lower than the current number of regions
 * while DAMON is running.  For such a case, repeat merging until the limit is
 * met while increasing @threshold up to possible maximum level.
 */
static void kdamond_merge_regions(struct damon_ctx *c, unsigned int threshold,
				  unsigned long sz_limit)
{
	struct damon_target *t;
	unsigned int nr_regions;
	unsigned int max_thres;

	max_thres = c->attrs.aggr_interval /
		(c->attrs.sample_interval ?  c->attrs.sample_interval : 1);
	do {
		nr_regions = 0;
		damon_for_each_target(t, c) {
			damon_merge_regions_of(t, threshold, sz_limit);
			nr_regions += damon_nr_regions(t);
		}
		threshold = max(1, threshold * 2);
	} while (nr_regions > c->attrs.max_nr_regions &&
			threshold / 2 < max_thres);
}

#ifdef CONFIG_DAMON_DEBUG_SANITY
static void damon_verify_split_region_at(struct damon_region *r,
		unsigned long sz_r)
{
	WARN_ONCE(sz_r == 0 || sz_r >= damon_sz_region(r),
			"sz_r: %lu r: %lu-%lu (%lu)\n",
			sz_r, r->ar.start, r->ar.end, damon_sz_region(r));
}
#else
static void damon_verify_split_region_at(struct damon_region *r,
		unsigned long sz_r)
{
}
#endif

/*
 * Split a region in two
 *
 * r		the region to be split
 * sz_r		size of the first sub-region that will be made
 */
static void damon_split_region_at(struct damon_target *t,
				  struct damon_region *r, unsigned long sz_r)
{
	struct damon_region *new;

	damon_verify_split_region_at(r, sz_r);
	new = damon_new_region(r->ar.start + sz_r, r->ar.end);
	if (!new)
		return;

	r->ar.end = new->ar.start;

	new->age = r->age;
	new->last_nr_accesses = r->last_nr_accesses;
	new->nr_accesses_bp = r->nr_accesses_bp;
	new->nr_accesses = r->nr_accesses;
	/* todo: do this for only installed probes */
	memcpy(new->probe_hits, r->probe_hits, sizeof(r->probe_hits));

	damon_insert_region(new, r, damon_next_region(r), t);
}

/* Split every region in the given target into 'nr_subs' regions */
static void damon_split_regions_of(struct damon_ctx *ctx,
				   struct damon_target *t, int nr_subs,
				   unsigned long min_region_sz)
{
	struct damon_region *r, *next;
	unsigned long sz_region, sz_sub = 0;
	int i;

	damon_for_each_region_safe(r, next, t) {
		sz_region = damon_sz_region(r);

		for (i = 0; i < nr_subs - 1 &&
				sz_region > 2 * min_region_sz; i++) {
			/*
			 * Randomly select size of left sub-region to be at
			 * least 10 percent and at most 90% of original region
			 */
			sz_sub = ALIGN_DOWN(damon_rand(ctx, 1, 10) *
					sz_region / 10, min_region_sz);
			/* Do not allow blank region */
			if (sz_sub == 0 || sz_sub >= sz_region)
				continue;

			damon_split_region_at(t, r, sz_sub);
			sz_region = sz_sub;
		}
	}
}

/*
 * Split every target region into randomly-sized small regions
 *
 * This function splits every target region into random-sized small regions if
 * current total number of the regions is equal or smaller than half of the
 * user-specified maximum number of regions.  This is for maximizing the
 * monitoring accuracy under the dynamically changeable access patterns.  If a
 * split was unnecessarily made, later 'kdamond_merge_regions()' will revert
 * it.
 */
static void kdamond_split_regions(struct damon_ctx *ctx)
{
	struct damon_target *t;
	unsigned int nr_regions = 0;
	static unsigned int last_nr_regions;
	int nr_subregions = 2;

	damon_for_each_target(t, ctx)
		nr_regions += damon_nr_regions(t);

	if (nr_regions > ctx->attrs.max_nr_regions / 2)
		return;

	/* Maybe the middle of the region has different access frequency */
	if (last_nr_regions == nr_regions &&
			nr_regions < ctx->attrs.max_nr_regions / 3)
		nr_subregions = 3;

	damon_for_each_target(t, ctx)
		damon_split_regions_of(ctx, t, nr_subregions,
				       ctx->min_region_sz);

	last_nr_regions = nr_regions;
}

/*
 * Check whether current monitoring should be stopped
 *
 * The monitoring is stopped when either the user requested to stop, or all
 * monitoring targets are invalid.
 *
 * Returns true if need to stop current monitoring.
 */
static bool kdamond_need_stop(struct damon_ctx *ctx)
{
	struct damon_target *t;

	if (kthread_should_stop())
		return true;

	if (!ctx->ops.target_valid)
		return false;

	damon_for_each_target(t, ctx) {
		if (ctx->ops.target_valid(t))
			return false;
	}

	return true;
}

static int damos_get_wmark_metric_value(enum damos_wmark_metric metric,
					unsigned long *metric_value)
{
	switch (metric) {
	case DAMOS_WMARK_FREE_MEM_RATE:
		*metric_value = global_zone_page_state(NR_FREE_PAGES) * 1000 /
		       totalram_pages();
		return 0;
	default:
		break;
	}
	return -EINVAL;
}

/*
 * Returns zero if the scheme is active.  Else, returns time to wait for next
 * watermark check in micro-seconds.
 */
static unsigned long damos_wmark_wait_us(struct damos *scheme)
{
	unsigned long metric;

	if (damos_get_wmark_metric_value(scheme->wmarks.metric, &metric))
		return 0;

	/* higher than high watermark or lower than low watermark */
	if (metric > scheme->wmarks.high || scheme->wmarks.low > metric) {
		if (scheme->wmarks.activated)
			pr_debug("deactivate a scheme (%d) for %s wmark\n",
				 scheme->action,
				 str_high_low(metric > scheme->wmarks.high));
		scheme->wmarks.activated = false;
		return scheme->wmarks.interval;
	}

	/* inactive and higher than middle watermark */
	if ((scheme->wmarks.high >= metric && metric >= scheme->wmarks.mid) &&
			!scheme->wmarks.activated)
		return scheme->wmarks.interval;

	if (!scheme->wmarks.activated)
		pr_debug("activate a scheme (%d)\n", scheme->action);
	scheme->wmarks.activated = true;
	return 0;
}

static void kdamond_usleep(unsigned long usecs)
{
	if (usecs >= USLEEP_RANGE_UPPER_BOUND)
		schedule_timeout_idle(usecs_to_jiffies(usecs));
	else
		usleep_range_idle(usecs, usecs + 1);
}

#ifdef CONFIG_DAMON_DEBUG_SANITY
static void damon_verify_ctx(struct damon_ctx *c)
{
	struct damon_target *t;
	struct damon_region *r;

	damon_for_each_target(t, c) {
		struct damon_region *prev_r = NULL;
		unsigned int nr_regions = 0;

		damon_for_each_region(r, t) {
			WARN_ONCE(r->ar.start >= r->ar.end,
					"region start (%lu) >= end (%lu)\n",
					r->ar.start, r->ar.end);
			WARN_ONCE(prev_r && prev_r->ar.end > r->ar.start,
					"region overlap (%lu > %lu)\n",
					prev_r->ar.end, r->ar.start);
			prev_r = r;
			nr_regions++;
		}
		WARN_ONCE(damon_nr_regions(t) != nr_regions,
				"nr_regions mismatch: %u != %u\n",
				damon_nr_regions(t), nr_regions);
	}
}
#else
static void damon_verify_ctx(struct damon_ctx *c)
{
}
#endif

/*
 * kdamond_call() - handle damon_call_control objects.
 * @ctx:	The &struct damon_ctx of the kdamond.
 * @cancel:	Whether to cancel the invocation of the function.
 *
 * If there are &struct damon_call_control requests that registered via
 * &damon_call() on @ctx, do or cancel the invocation of the function depending
 * on @cancel.  @cancel is set when the kdamond is already out of the main loop
 * and therefore will be terminated.
 */
static void kdamond_call(struct damon_ctx *ctx, bool cancel)
{
	struct damon_call_control *control, *next;
	LIST_HEAD(controls);

	damon_verify_ctx(ctx);

	mutex_lock(&ctx->call_controls_lock);
	list_splice_tail_init(&ctx->call_controls, &controls);
	mutex_unlock(&ctx->call_controls_lock);

	list_for_each_entry_safe(control, next, &controls, list) {
		if (!control->repeat || cancel)
			list_del(&control->list);

		if (cancel)
			control->canceled = true;
		else
			control->return_code = control->fn(control->data);

		if (!control->repeat)
			complete(&control->completion);
		else if (control->canceled && control->dealloc_on_cancel)
			kfree(control);
		if (!cancel && ctx->maybe_corrupted)
			break;
	}

	mutex_lock(&ctx->call_controls_lock);
	list_splice_tail(&controls, &ctx->call_controls);
	mutex_unlock(&ctx->call_controls_lock);
}

/* Returns negative error code if it's not activated but should return */
static int kdamond_wait_activation(struct damon_ctx *ctx)
{
	struct damos *s;
	unsigned long wait_time;
	unsigned long min_wait_time = 0;
	bool init_wait_time = false;

	while (!kdamond_need_stop(ctx)) {
		damon_for_each_scheme(s, ctx) {
			wait_time = damos_wmark_wait_us(s);
			if (!init_wait_time || wait_time < min_wait_time) {
				init_wait_time = true;
				min_wait_time = wait_time;
			}
		}
		if (!min_wait_time)
			return 0;

		kdamond_usleep(min_wait_time);

		kdamond_call(ctx, false);
		if (ctx->maybe_corrupted)
			return -EINVAL;
		damos_walk_cancel(ctx);
	}
	return -EBUSY;
}

static void kdamond_init_ctx(struct damon_ctx *ctx)
{
	unsigned long sample_interval = ctx->attrs.sample_interval ?
		ctx->attrs.sample_interval : 1;
	struct damos *scheme;

	ctx->passed_sample_intervals = 0;
	ctx->next_aggregation_sis = ctx->attrs.aggr_interval / sample_interval;
	ctx->next_ops_update_sis = ctx->attrs.ops_update_interval /
		sample_interval;
	ctx->next_intervals_tune_sis = ctx->next_aggregation_sis *
		ctx->attrs.intervals_goal.aggrs;

	damon_for_each_scheme(scheme, ctx) {
		damos_set_next_apply_sis(scheme, ctx);
		damos_set_filters_default_reject(scheme);
	}
}

/*
 * The monitoring daemon that runs as a kernel thread
 */
static int kdamond_fn(void *data)
{
	struct damon_ctx *ctx = data;
	unsigned int max_nr_accesses = 0;
	unsigned long sz_limit = 0;

	pr_debug("kdamond (%d) starts\n", current->pid);

	mutex_lock(&ctx->call_controls_lock);
	ctx->call_controls_obsolete = false;
	mutex_unlock(&ctx->call_controls_lock);
	mutex_lock(&ctx->walk_control_lock);
	ctx->walk_control_obsolete = false;
	mutex_unlock(&ctx->walk_control_lock);
	complete(&ctx->kdamond_started);
	kdamond_init_ctx(ctx);

	if (ctx->ops.init)
		ctx->ops.init(ctx);
	ctx->regions_score_histogram = kmalloc_array(DAMOS_MAX_SCORE + 1,
			sizeof(*ctx->regions_score_histogram), GFP_KERNEL);
	if (!ctx->regions_score_histogram)
		goto done;

	sz_limit = damon_apply_min_nr_regions(ctx);

	while (!kdamond_need_stop(ctx)) {
		/*
		 * ctx->attrs and ctx->next_{aggregation,ops_update}_sis could
		 * be changed from kdamond_call().  Read the values here, and
		 * use those for this iteration.  That is, damon_set_attrs()
		 * updated new values are respected from next iteration.
		 */
		unsigned long next_aggregation_sis = ctx->next_aggregation_sis;
		unsigned long next_ops_update_sis = ctx->next_ops_update_sis;
		unsigned long sample_interval = ctx->attrs.sample_interval;

		if (kdamond_wait_activation(ctx))
			break;

		if (ctx->ops.prepare_access_checks)
			ctx->ops.prepare_access_checks(ctx);

		kdamond_usleep(sample_interval);
		ctx->passed_sample_intervals++;

		if (ctx->ops.check_accesses)
			max_nr_accesses = ctx->ops.check_accesses(ctx);
		if (ctx->ops.apply_probes)
			ctx->ops.apply_probes(ctx);

		if (time_after_eq(ctx->passed_sample_intervals,
					next_aggregation_sis)) {
			kdamond_merge_regions(ctx,
					max_nr_accesses / 10,
					sz_limit);
			/* online updates might be made */
			sz_limit = damon_apply_min_nr_regions(ctx);
		}

		/*
		 * do kdamond_call() and kdamond_apply_schemes() after
		 * kdamond_merge_regions() if possible, to reduce overhead
		 */
		kdamond_call(ctx, false);
		if (ctx->maybe_corrupted)
			break;
		while (ctx->pause) {
			damos_walk_cancel(ctx);
			kdamond_usleep(ctx->attrs.sample_interval);
			/* allow caller unset pause via damon_call() */
			kdamond_call(ctx, false);
			if (kdamond_need_stop(ctx) || ctx->maybe_corrupted)
				goto done;
		}
		if (!list_empty(&ctx->schemes))
			kdamond_apply_schemes(ctx);
		else
			damos_walk_cancel(ctx);

		sample_interval = ctx->attrs.sample_interval ?
			ctx->attrs.sample_interval : 1;
		if (time_after_eq(ctx->passed_sample_intervals,
					next_aggregation_sis)) {
			if (ctx->attrs.intervals_goal.aggrs &&
					time_after_eq(
						ctx->passed_sample_intervals,
						ctx->next_intervals_tune_sis)) {
				/*
				 * ctx->next_aggregation_sis might be updated
				 * from kdamond_call().  In the case,
				 * damon_set_attrs() which will be called from
				 * kdamond_tune_interval() may wrongly think
				 * this is in the middle of the current
				 * aggregation, and make aggregation
				 * information reset for all regions.  Then,
				 * following kdamond_reset_aggregated() call
				 * will make the region information invalid,
				 * particularly for ->nr_accesses_bp.
				 *
				 * Reset ->next_aggregation_sis to avoid that.
				 * It will anyway correctly updated after this
				 * if clause.
				 */
				ctx->next_aggregation_sis =
					next_aggregation_sis;
				ctx->next_intervals_tune_sis +=
					ctx->attrs.aggr_samples *
					ctx->attrs.intervals_goal.aggrs;
				kdamond_tune_intervals(ctx);
				sample_interval = ctx->attrs.sample_interval ?
					ctx->attrs.sample_interval : 1;

			}
			ctx->next_aggregation_sis = next_aggregation_sis +
				ctx->attrs.aggr_interval / sample_interval;

			kdamond_reset_aggregated(ctx);
			kdamond_split_regions(ctx);
		}

		if (time_after_eq(ctx->passed_sample_intervals,
					next_ops_update_sis)) {
			ctx->next_ops_update_sis = next_ops_update_sis +
				ctx->attrs.ops_update_interval /
				sample_interval;
			if (ctx->ops.update)
				ctx->ops.update(ctx);
		}
	}
done:
	damon_destroy_targets(ctx);

	kfree(ctx->regions_score_histogram);
	mutex_lock(&ctx->call_controls_lock);
	ctx->call_controls_obsolete = true;
	mutex_unlock(&ctx->call_controls_lock);
	kdamond_call(ctx, true);
	mutex_lock(&ctx->walk_control_lock);
	ctx->walk_control_obsolete = true;
	mutex_unlock(&ctx->walk_control_lock);
	damos_walk_cancel(ctx);

	pr_debug("kdamond (%d) finishes\n", current->pid);
	mutex_lock(&ctx->kdamond_lock);
	ctx->kdamond = NULL;
	mutex_unlock(&ctx->kdamond_lock);

	mutex_lock(&damon_lock);
	nr_running_ctxs--;
	if (!nr_running_ctxs && running_exclusive_ctxs)
		running_exclusive_ctxs = false;
	mutex_unlock(&damon_lock);

	return 0;
}

struct damon_system_ram_range_walk_arg {
	bool walked;
	struct resource res;
};

static int damon_system_ram_walk_fn(struct resource *res, void *arg)
{
	struct damon_system_ram_range_walk_arg *a = arg;

	if (!a->walked) {
		a->walked = true;
		a->res.start = res->start;
	}
	a->res.end = res->end;
	return 0;
}

static unsigned long damon_res_to_core_addr(resource_size_t ra,
		unsigned long addr_unit)
{
	/*
	 * Use div_u64() for avoiding linking errors related with __udivdi3,
	 * __aeabi_uldivmod, or similar problems.  This should also improve the
	 * performance optimization (read div_u64() comment for the detail).
	 */
	if (sizeof(ra) == 8 && sizeof(addr_unit) == 4)
		return div_u64(ra, addr_unit);
	return ra / addr_unit;
}

static bool damon_find_system_rams_range(unsigned long *start,
		unsigned long *end, unsigned long addr_unit)
{
	struct damon_system_ram_range_walk_arg arg = {};

	walk_system_ram_res(0, -1, &arg, damon_system_ram_walk_fn);
	if (!arg.walked)
		return false;
	*start = damon_res_to_core_addr(arg.res.start, addr_unit);
	*end = damon_res_to_core_addr(arg.res.end + 1, addr_unit);
	if (*end <= *start)
		return false;
	return true;
}

/**
 * damon_set_region_system_rams_default() - Set the region of the given
 * monitoring target as requested, or to cover all 'System RAM' resources.
 * @t:		The monitoring target to set the region.
 * @start:	The pointer to the start address of the region.
 * @end:	The pointer to the end address of the region.
 * @addr_unit:	The address unit for the damon_ctx of @t.
 * @min_region_sz:	Minimum region size.
 *
 * This function sets the region of @t as requested by @start and @end.  If the
 * values of @start and @end are zero, however, this function finds 'System
 * RAM' resources and sets the region to cover all the resource.  In the latter
 * case, this function saves the start and the end addresseses of the first and
 * the last resources in @start and @end, respectively.
 *
 * Return: 0 on success, negative error code otherwise.
 */
int damon_set_region_system_rams_default(struct damon_target *t,
			unsigned long *start, unsigned long *end,
			unsigned long addr_unit, unsigned long min_region_sz)
{
	struct damon_addr_range addr_range;

	if (*start > *end)
		return -EINVAL;

	if (!*start && !*end &&
		!damon_find_system_rams_range(start, end, addr_unit))
		return -EINVAL;

	addr_range.start = *start;
	addr_range.end = *end;
	return damon_set_regions(t, &addr_range, 1, min_region_sz);
}

/*
 * damon_moving_sum() - Calculate an inferred moving sum value.
 * @mvsum:	Inferred sum of the last @len_window values.
 * @nomvsum:	Non-moving sum of the last discrete @len_window window values.
 * @len_window:	The number of last values to take care of.
 * @new_value:	New value that will be added to the pseudo moving sum.
 *
 * Moving sum (moving average * window size) is good for handling noise, but
 * the cost of keeping past values can be high for arbitrary window size.  This
 * function implements a lightweight pseudo moving sum function that doesn't
 * keep the past window values.
 *
 * It simply assumes there was no noise in the past, and get the no-noise
 * assumed past value to drop from @nomvsum and @len_window.  @nomvsum is a
 * non-moving sum of the last window.  For example, if @len_window is 10 and we
 * have 25 values, @nomvsum is the sum of the 11th to 20th values of the 25
 * values.  Hence, this function simply drops @nomvsum / @len_window from
 * given @mvsum and add @new_value.
 *
 * For example, if @len_window is 10 and @nomvsum is 50, the last 10 values for
 * the last window could be vary, e.g., 0, 10, 0, 10, 0, 10, 0, 0, 0, 20.  For
 * calculating next moving sum with a new value, we should drop 0 from 50 and
 * add the new value.  However, this function assumes it got value 5 for each
 * of the last ten times.  Based on the assumption, when the next value is
 * measured, it drops the assumed past value, 5 from the current sum, and add
 * the new value to get the updated pseduo-moving average.
 *
 * This means the value could have errors, but the errors will be disappeared
 * for every @len_window aligned calls.  For example, if @len_window is 10, the
 * pseudo moving sum with 11th value to 19th value would have an error.  But
 * the sum with 20th value will not have the error.
 *
 * Return: Pseudo-moving average after getting the @new_value.
 */
static unsigned int damon_moving_sum(unsigned int mvsum, unsigned int nomvsum,
		unsigned int len_window, unsigned int new_value)
{
	return mvsum - nomvsum / len_window + new_value;
}

/**
 * damon_update_region_access_rate() - Update the access rate of a region.
 * @r:		The DAMON region to update for its access check result.
 * @accessed:	Whether the region has accessed during last sampling interval.
 * @attrs:	The damon_attrs of the DAMON context.
 *
 * Update the access rate of a region with the region's last sampling interval
 * access check result.
 *
 * Usually this will be called by &damon_operations->check_accesses callback.
 */
void damon_update_region_access_rate(struct damon_region *r, bool accessed,
		struct damon_attrs *attrs)
{
	unsigned int len_window = 1;

	/*
	 * sample_interval can be zero, but cannot be larger than
	 * aggr_interval, owing to validation of damon_set_attrs().
	 */
	if (attrs->sample_interval)
		len_window = damon_max_nr_accesses(attrs);
	r->nr_accesses_bp = damon_moving_sum(r->nr_accesses_bp,
			r->last_nr_accesses * 10000, len_window,
			accessed ? 10000 : 0);

	if (accessed)
		r->nr_accesses++;
}

/**
 * damon_initialized() - Return if DAMON is ready to be used.
 *
 * Return: true if DAMON is ready to be used, false otherwise.
 */
bool damon_initialized(void)
{
	return damon_region_cache != NULL;
}

static int __init damon_init(void)
{
	damon_region_cache = KMEM_CACHE(damon_region, 0);
	if (unlikely(!damon_region_cache)) {
		pr_err("creating damon_region_cache fails\n");
		return -ENOMEM;
	}

	return 0;
}

subsys_initcall(damon_init);

#include "tests/core-kunit.h"