cgroup.c

	 * while_each_thread(). Besides, a forking task that has passed
	 * cgroup_post_fork() without seeing use_task_css_set_links = 1
	 * is not guaranteed to have its child immediately visible in the
	 * tasklist if we walk through it with RCU.
	 */
	read_lock(&tasklist_lock);
	do_each_thread(g, p) {
		task_lock(p);
		/*
		 * We should check if the process is exiting, otherwise
		 * it will race with cgroup_exit() in that the list
		 * entry won't be deleted though the process has exited.
		 */
		if (!(p->flags & PF_EXITING) && list_empty(&p->cg_list))
			list_add(&p->cg_list, &task_css_set(p)->tasks);
		task_unlock(p);
	} while_each_thread(g, p);
	read_unlock(&tasklist_lock);
	write_unlock(&css_set_lock);
}

/**
 * css_next_child - find the next child of a given css
 * @pos_css: the current position (%NULL to initiate traversal)
 * @parent_css: css whose children to walk
 *
 * This function returns the next child of @parent_css and should be called
 * under RCU read lock.  The only requirement is that @parent_css and
 * @pos_css are accessible.  The next sibling is guaranteed to be returned
 * regardless of their states.
 */
struct cgroup_subsys_state *
css_next_child(struct cgroup_subsys_state *pos_css,
	       struct cgroup_subsys_state *parent_css)
{
	struct cgroup *pos = pos_css ? pos_css->cgroup : NULL;
	struct cgroup *cgrp = parent_css->cgroup;
	struct cgroup *next;

	WARN_ON_ONCE(!rcu_read_lock_held());

	/*
	 * @pos could already have been removed.  Once a cgroup is removed,
	 * its ->sibling.next is no longer updated when its next sibling
	 * changes.  As CGRP_DEAD assertion is serialized and happens
	 * before the cgroup is taken off the ->sibling list, if we see it
	 * unasserted, it's guaranteed that the next sibling hasn't
	 * finished its grace period even if it's already removed, and thus
	 * safe to dereference from this RCU critical section.  If
	 * ->sibling.next is inaccessible, cgroup_is_dead() is guaranteed
	 * to be visible as %true here.
	 *
	 * If @pos is dead, its next pointer can't be dereferenced;
	 * however, as each cgroup is given a monotonically increasing
	 * unique serial number and always appended to the sibling list,
	 * the next one can be found by walking the parent's children until
	 * we see a cgroup with higher serial number than @pos's.  While
	 * this path can be slower, it's taken only when either the current
	 * cgroup is removed or iteration and removal race.
	 */
	if (!pos) {
		next = list_entry_rcu(cgrp->children.next, struct cgroup, sibling);
	} else if (likely(!cgroup_is_dead(pos))) {
		next = list_entry_rcu(pos->sibling.next, struct cgroup, sibling);
	} else {
		list_for_each_entry_rcu(next, &cgrp->children, sibling)
			if (next->serial_nr > pos->serial_nr)
				break;
	}

	if (&next->sibling == &cgrp->children)
		return NULL;

	if (parent_css->ss)
		return cgroup_css(next, parent_css->ss->subsys_id);
	else
		return &next->dummy_css;
}
EXPORT_SYMBOL_GPL(css_next_child);

/**
 * css_next_descendant_pre - find the next descendant for pre-order walk
 * @pos: the current position (%NULL to initiate traversal)
 * @root: css whose descendants to walk
 *
 * To be used by css_for_each_descendant_pre().  Find the next descendant
 * to visit for pre-order traversal of @root's descendants.
 *
 * While this function requires RCU read locking, it doesn't require the
 * whole traversal to be contained in a single RCU critical section.  This
 * function will return the correct next descendant as long as both @pos
 * and @root are accessible and @pos is a descendant of @root.
 */
struct cgroup_subsys_state *
css_next_descendant_pre(struct cgroup_subsys_state *pos,
			struct cgroup_subsys_state *root)
{
	struct cgroup_subsys_state *next;

	WARN_ON_ONCE(!rcu_read_lock_held());

	/* if first iteration, pretend we just visited @root */
	if (!pos)
		pos = root;

	/* visit the first child if exists */
	next = css_next_child(NULL, pos);
	if (next)
		return next;

	/* no child, visit my or the closest ancestor's next sibling */
	while (pos != root) {
		next = css_next_child(pos, css_parent(pos));
		if (next)
			return next;
		pos = css_parent(pos);
	}

	return NULL;
}
EXPORT_SYMBOL_GPL(css_next_descendant_pre);

/**
 * css_rightmost_descendant - return the rightmost descendant of a css
 * @pos: css of interest
 *
 * Return the rightmost descendant of @pos.  If there's no descendant, @pos
 * is returned.  This can be used during pre-order traversal to skip
 * subtree of @pos.
 *
 * While this function requires RCU read locking, it doesn't require the
 * whole traversal to be contained in a single RCU critical section.  This
 * function will return the correct rightmost descendant as long as @pos is
 * accessible.
 */
struct cgroup_subsys_state *
css_rightmost_descendant(struct cgroup_subsys_state *pos)
{
	struct cgroup_subsys_state *last, *tmp;

	WARN_ON_ONCE(!rcu_read_lock_held());

	do {
		last = pos;
		/* ->prev isn't RCU safe, walk ->next till the end */
		pos = NULL;
		css_for_each_child(tmp, last)
			pos = tmp;
	} while (pos);

	return last;
}
EXPORT_SYMBOL_GPL(css_rightmost_descendant);

static struct cgroup_subsys_state *
css_leftmost_descendant(struct cgroup_subsys_state *pos)
{
	struct cgroup_subsys_state *last;

	do {
		last = pos;
		pos = css_next_child(NULL, pos);
	} while (pos);

	return last;
}

/**
 * css_next_descendant_post - find the next descendant for post-order walk
 * @pos: the current position (%NULL to initiate traversal)
 * @root: css whose descendants to walk
 *
 * To be used by css_for_each_descendant_post().  Find the next descendant
 * to visit for post-order traversal of @root's descendants.
 *
 * While this function requires RCU read locking, it doesn't require the
 * whole traversal to be contained in a single RCU critical section.  This
 * function will return the correct next descendant as long as both @pos
 * and @cgroup are accessible and @pos is a descendant of @cgroup.
 */
struct cgroup_subsys_state *
css_next_descendant_post(struct cgroup_subsys_state *pos,
			 struct cgroup_subsys_state *root)
{
	struct cgroup_subsys_state *next;

	WARN_ON_ONCE(!rcu_read_lock_held());

	/* if first iteration, visit the leftmost descendant */
	if (!pos) {
		next = css_leftmost_descendant(root);
		return next != root ? next : NULL;
	}

	/* if there's an unvisited sibling, visit its leftmost descendant */
	next = css_next_child(pos, css_parent(pos));
	if (next)
		return css_leftmost_descendant(next);

	/* no sibling left, visit parent */
	next = css_parent(pos);
	return next != root ? next : NULL;
}
EXPORT_SYMBOL_GPL(css_next_descendant_post);

/**
 * cgroup_advance_task_iter - advance a task itererator to the next css_set
 * @cgrp: the cgroup to walk tasks of
 * @it: the iterator to advance
 *
 * Advance @it to the next css_set to walk.
 */
static void cgroup_advance_task_iter(struct cgroup *cgrp,
				     struct cgroup_task_iter *it)
{
	struct list_head *l = it->cset_link;
	struct cgrp_cset_link *link;
	struct css_set *cset;

	/* Advance to the next non-empty css_set */
	do {
		l = l->next;
		if (l == &cgrp->cset_links) {
			it->cset_link = NULL;
			return;
		}
		link = list_entry(l, struct cgrp_cset_link, cset_link);
		cset = link->cset;
	} while (list_empty(&cset->tasks));
	it->cset_link = l;
	it->task = cset->tasks.next;
}

/**
 * cgroup_task_iter_start - initiate task iteration
 * @cgrp: the cgroup to walk tasks of
 * @it: the task iterator to use
 *
 * Initiate iteration through the tasks of @cgrp.  The caller can call
 * cgroup_task_iter_next() to walk through the tasks until the function
 * returns NULL.  On completion of iteration, cgroup_task_iter_end() must
 * be called.
 *
 * Note that this function acquires a lock which is released when the
 * iteration finishes.  The caller can't sleep while iteration is in
 * progress.
 */
void cgroup_task_iter_start(struct cgroup *cgrp, struct cgroup_task_iter *it)
	__acquires(css_set_lock)
{
	/*
	 * The first time anyone tries to iterate across a cgroup,
	 * we need to enable the list linking each css_set to its
	 * tasks, and fix up all existing tasks.
	 */
	if (!use_task_css_set_links)
		cgroup_enable_task_cg_lists();

	read_lock(&css_set_lock);
	it->cset_link = &cgrp->cset_links;
	cgroup_advance_task_iter(cgrp, it);
}

/**
 * cgroup_task_iter_next - return the next task for the iterator
 * @cgrp: the cgroup to walk tasks of
 * @it: the task iterator being iterated
 *
 * The "next" function for task iteration.  @it should have been
 * initialized via cgroup_task_iter_start().  Returns NULL when the
 * iteration reaches the end.
 */
struct task_struct *cgroup_task_iter_next(struct cgroup *cgrp,
					  struct cgroup_task_iter *it)
{
	struct task_struct *res;
	struct list_head *l = it->task;
	struct cgrp_cset_link *link;

	/* If the iterator cg is NULL, we have no tasks */
	if (!it->cset_link)
		return NULL;
	res = list_entry(l, struct task_struct, cg_list);
	/* Advance iterator to find next entry */
	l = l->next;
	link = list_entry(it->cset_link, struct cgrp_cset_link, cset_link);
	if (l == &link->cset->tasks) {
		/*
		 * We reached the end of this task list - move on to the
		 * next cgrp_cset_link.
		 */
		cgroup_advance_task_iter(cgrp, it);
	} else {
		it->task = l;
	}
	return res;
}

/**
 * cgroup_task_iter_end - finish task iteration
 * @cgrp: the cgroup to walk tasks of
 * @it: the task iterator to finish
 *
 * Finish task iteration started by cgroup_task_iter_start().
 */
void cgroup_task_iter_end(struct cgroup *cgrp, struct cgroup_task_iter *it)
	__releases(css_set_lock)
{
	read_unlock(&css_set_lock);
}

static inline int started_after_time(struct task_struct *t1,
				     struct timespec *time,
				     struct task_struct *t2)
{
	int start_diff = timespec_compare(&t1->start_time, time);
	if (start_diff > 0) {
		return 1;
	} else if (start_diff < 0) {
		return 0;
	} else {
		/*
		 * Arbitrarily, if two processes started at the same
		 * time, we'll say that the lower pointer value
		 * started first. Note that t2 may have exited by now
		 * so this may not be a valid pointer any longer, but
		 * that's fine - it still serves to distinguish
		 * between two tasks started (effectively) simultaneously.
		 */
		return t1 > t2;
	}
}

/*
 * This function is a callback from heap_insert() and is used to order
 * the heap.
 * In this case we order the heap in descending task start time.
 */
static inline int started_after(void *p1, void *p2)
{
	struct task_struct *t1 = p1;
	struct task_struct *t2 = p2;
	return started_after_time(t1, &t2->start_time, t2);
}

/**
 * cgroup_scan_tasks - iterate though all the tasks in a cgroup
 * @scan: struct cgroup_scanner containing arguments for the scan
 *
 * Arguments include pointers to callback functions test_task() and
 * process_task().
 * Iterate through all the tasks in a cgroup, calling test_task() for each,
 * and if it returns true, call process_task() for it also.
 * The test_task pointer may be NULL, meaning always true (select all tasks).
 * Effectively duplicates cgroup_task_iter_{start,next,end}()
 * but does not lock css_set_lock for the call to process_task().
 * The struct cgroup_scanner may be embedded in any structure of the caller's
 * creation.
 * It is guaranteed that process_task() will act on every task that
 * is a member of the cgroup for the duration of this call. This
 * function may or may not call process_task() for tasks that exit
 * or move to a different cgroup during the call, or are forked or
 * move into the cgroup during the call.
 *
 * Note that test_task() may be called with locks held, and may in some
 * situations be called multiple times for the same task, so it should
 * be cheap.
 * If the heap pointer in the struct cgroup_scanner is non-NULL, a heap has been
 * pre-allocated and will be used for heap operations (and its "gt" member will
 * be overwritten), else a temporary heap will be used (allocation of which
 * may cause this function to fail).
 */
int cgroup_scan_tasks(struct cgroup_scanner *scan)
{
	int retval, i;
	struct cgroup_task_iter it;
	struct task_struct *p, *dropped;
	/* Never dereference latest_task, since it's not refcounted */
	struct task_struct *latest_task = NULL;
	struct ptr_heap tmp_heap;
	struct ptr_heap *heap;
	struct timespec latest_time = { 0, 0 };

	if (scan->heap) {
		/* The caller supplied our heap and pre-allocated its memory */
		heap = scan->heap;
		heap->gt = &started_after;
	} else {
		/* We need to allocate our own heap memory */
		heap = &tmp_heap;
		retval = heap_init(heap, PAGE_SIZE, GFP_KERNEL, &started_after);
		if (retval)
			/* cannot allocate the heap */
			return retval;
	}

 again:
	/*
	 * Scan tasks in the cgroup, using the scanner's "test_task" callback
	 * to determine which are of interest, and using the scanner's
	 * "process_task" callback to process any of them that need an update.
	 * Since we don't want to hold any locks during the task updates,
	 * gather tasks to be processed in a heap structure.
	 * The heap is sorted by descending task start time.
	 * If the statically-sized heap fills up, we overflow tasks that
	 * started later, and in future iterations only consider tasks that
	 * started after the latest task in the previous pass. This
	 * guarantees forward progress and that we don't miss any tasks.
	 */
	heap->size = 0;
	cgroup_task_iter_start(scan->cgrp, &it);
	while ((p = cgroup_task_iter_next(scan->cgrp, &it))) {
		/*
		 * Only affect tasks that qualify per the caller's callback,
		 * if he provided one
		 */
		if (scan->test_task && !scan->test_task(p, scan))
			continue;
		/*
		 * Only process tasks that started after the last task
		 * we processed
		 */
		if (!started_after_time(p, &latest_time, latest_task))
			continue;
		dropped = heap_insert(heap, p);
		if (dropped == NULL) {
			/*
			 * The new task was inserted; the heap wasn't
			 * previously full
			 */
			get_task_struct(p);
		} else if (dropped != p) {
			/*
			 * The new task was inserted, and pushed out a
			 * different task
			 */
			get_task_struct(p);
			put_task_struct(dropped);
		}
		/*
		 * Else the new task was newer than anything already in
		 * the heap and wasn't inserted
		 */
	}
	cgroup_task_iter_end(scan->cgrp, &it);

	if (heap->size) {
		for (i = 0; i < heap->size; i++) {
			struct task_struct *q = heap->ptrs[i];
			if (i == 0) {
				latest_time = q->start_time;
				latest_task = q;
			}
			/* Process the task per the caller's callback */
			scan->process_task(q, scan);
			put_task_struct(q);
		}
		/*
		 * If we had to process any tasks at all, scan again
		 * in case some of them were in the middle of forking
		 * children that didn't get processed.
		 * Not the most efficient way to do it, but it avoids
		 * having to take callback_mutex in the fork path
		 */
		goto again;
	}
	if (heap == &tmp_heap)
		heap_free(&tmp_heap);
	return 0;
}

static void cgroup_transfer_one_task(struct task_struct *task,
				     struct cgroup_scanner *scan)
{
	struct cgroup *new_cgroup = scan->data;

	mutex_lock(&cgroup_mutex);
	cgroup_attach_task(new_cgroup, task, false);
	mutex_unlock(&cgroup_mutex);
}

/**
 * cgroup_trasnsfer_tasks - move tasks from one cgroup to another
 * @to: cgroup to which the tasks will be moved
 * @from: cgroup in which the tasks currently reside
 */
int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from)
{
	struct cgroup_scanner scan;

	scan.cgrp = from;
	scan.test_task = NULL; /* select all tasks in cgroup */
	scan.process_task = cgroup_transfer_one_task;
	scan.heap = NULL;
	scan.data = to;

	return cgroup_scan_tasks(&scan);
}

/*
 * Stuff for reading the 'tasks'/'procs' files.
 *
 * Reading this file can return large amounts of data if a cgroup has
 * *lots* of attached tasks. So it may need several calls to read(),
 * but we cannot guarantee that the information we produce is correct
 * unless we produce it entirely atomically.
 *
 */

/* which pidlist file are we talking about? */
enum cgroup_filetype {
	CGROUP_FILE_PROCS,
	CGROUP_FILE_TASKS,
};

/*
 * A pidlist is a list of pids that virtually represents the contents of one
 * of the cgroup files ("procs" or "tasks"). We keep a list of such pidlists,
 * a pair (one each for procs, tasks) for each pid namespace that's relevant
 * to the cgroup.
 */
struct cgroup_pidlist {
	/*
	 * used to find which pidlist is wanted. doesn't change as long as
	 * this particular list stays in the list.
	*/
	struct { enum cgroup_filetype type; struct pid_namespace *ns; } key;
	/* array of xids */
	pid_t *list;
	/* how many elements the above list has */
	int length;
	/* how many files are using the current array */
	int use_count;
	/* each of these stored in a list by its cgroup */
	struct list_head links;
	/* pointer to the cgroup we belong to, for list removal purposes */
	struct cgroup *owner;
	/* protects the other fields */
	struct rw_semaphore rwsem;
};

/*
 * The following two functions "fix" the issue where there are more pids
 * than kmalloc will give memory for; in such cases, we use vmalloc/vfree.
 * TODO: replace with a kernel-wide solution to this problem
 */
#define PIDLIST_TOO_LARGE(c) ((c) * sizeof(pid_t) > (PAGE_SIZE * 2))
static void *pidlist_allocate(int count)
{
	if (PIDLIST_TOO_LARGE(count))
		return vmalloc(count * sizeof(pid_t));
	else
		return kmalloc(count * sizeof(pid_t), GFP_KERNEL);
}
static void pidlist_free(void *p)
{
	if (is_vmalloc_addr(p))
		vfree(p);
	else
		kfree(p);
}

/*
 * pidlist_uniq - given a kmalloc()ed list, strip out all duplicate entries
 * Returns the number of unique elements.
 */
static int pidlist_uniq(pid_t *list, int length)
{
	int src, dest = 1;

	/*
	 * we presume the 0th element is unique, so i starts at 1. trivial
	 * edge cases first; no work needs to be done for either
	 */
	if (length == 0 || length == 1)
		return length;
	/* src and dest walk down the list; dest counts unique elements */
	for (src = 1; src < length; src++) {
		/* find next unique element */
		while (list[src] == list[src-1]) {
			src++;
			if (src == length)
				goto after;
		}
		/* dest always points to where the next unique element goes */
		list[dest] = list[src];
		dest++;
	}
after:
	return dest;
}

static int cmppid(const void *a, const void *b)
{
	return *(pid_t *)a - *(pid_t *)b;
}

/*
 * find the appropriate pidlist for our purpose (given procs vs tasks)
 * returns with the lock on that pidlist already held, and takes care
 * of the use count, or returns NULL with no locks held if we're out of
 * memory.
 */
static struct cgroup_pidlist *cgroup_pidlist_find(struct cgroup *cgrp,
						  enum cgroup_filetype type)
{
	struct cgroup_pidlist *l;
	/* don't need task_nsproxy() if we're looking at ourself */
	struct pid_namespace *ns = task_active_pid_ns(current);

	/*
	 * We can't drop the pidlist_mutex before taking the l->rwsem in case
	 * the last ref-holder is trying to remove l from the list at the same
	 * time. Holding the pidlist_mutex precludes somebody taking whichever
	 * list we find out from under us - compare release_pid_array().
	 */
	mutex_lock(&cgrp->pidlist_mutex);
	list_for_each_entry(l, &cgrp->pidlists, links) {
		if (l->key.type == type && l->key.ns == ns) {
			/* make sure l doesn't vanish out from under us */
			down_write(&l->rwsem);
			mutex_unlock(&cgrp->pidlist_mutex);
			return l;
		}
	}
	/* entry not found; create a new one */
	l = kzalloc(sizeof(struct cgroup_pidlist), GFP_KERNEL);
	if (!l) {
		mutex_unlock(&cgrp->pidlist_mutex);
		return l;
	}
	init_rwsem(&l->rwsem);
	down_write(&l->rwsem);
	l->key.type = type;
	l->key.ns = get_pid_ns(ns);
	l->owner = cgrp;
	list_add(&l->links, &cgrp->pidlists);
	mutex_unlock(&cgrp->pidlist_mutex);
	return l;
}

/*
 * Load a cgroup's pidarray with either procs' tgids or tasks' pids
 */
static int pidlist_array_load(struct cgroup *cgrp, enum cgroup_filetype type,
			      struct cgroup_pidlist **lp)
{
	pid_t *array;
	int length;
	int pid, n = 0; /* used for populating the array */
	struct cgroup_task_iter it;
	struct task_struct *tsk;
	struct cgroup_pidlist *l;

	/*
	 * If cgroup gets more users after we read count, we won't have
	 * enough space - tough.  This race is indistinguishable to the
	 * caller from the case that the additional cgroup users didn't
	 * show up until sometime later on.
	 */
	length = cgroup_task_count(cgrp);
	array = pidlist_allocate(length);
	if (!array)
		return -ENOMEM;
	/* now, populate the array */
	cgroup_task_iter_start(cgrp, &it);
	while ((tsk = cgroup_task_iter_next(cgrp, &it))) {
		if (unlikely(n == length))
			break;
		/* get tgid or pid for procs or tasks file respectively */
		if (type == CGROUP_FILE_PROCS)
			pid = task_tgid_vnr(tsk);
		else
			pid = task_pid_vnr(tsk);
		if (pid > 0) /* make sure to only use valid results */
			array[n++] = pid;
	}
	cgroup_task_iter_end(cgrp, &it);
	length = n;
	/* now sort & (if procs) strip out duplicates */
	sort(array, length, sizeof(pid_t), cmppid, NULL);
	if (type == CGROUP_FILE_PROCS)
		length = pidlist_uniq(array, length);
	l = cgroup_pidlist_find(cgrp, type);
	if (!l) {
		pidlist_free(array);
		return -ENOMEM;
	}
	/* store array, freeing old if necessary - lock already held */
	pidlist_free(l->list);
	l->list = array;
	l->length = length;
	l->use_count++;
	up_write(&l->rwsem);
	*lp = l;
	return 0;
}

/**
 * cgroupstats_build - build and fill cgroupstats
 * @stats: cgroupstats to fill information into
 * @dentry: A dentry entry belonging to the cgroup for which stats have
 * been requested.
 *
 * Build and fill cgroupstats so that taskstats can export it to user
 * space.
 */
int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry)
{
	int ret = -EINVAL;
	struct cgroup *cgrp;
	struct cgroup_task_iter it;
	struct task_struct *tsk;

	/*
	 * Validate dentry by checking the superblock operations,
	 * and make sure it's a directory.
	 */
	if (dentry->d_sb->s_op != &cgroup_ops ||
	    !S_ISDIR(dentry->d_inode->i_mode))
		 goto err;

	ret = 0;
	cgrp = dentry->d_fsdata;

	cgroup_task_iter_start(cgrp, &it);
	while ((tsk = cgroup_task_iter_next(cgrp, &it))) {
		switch (tsk->state) {
		case TASK_RUNNING:
			stats->nr_running++;
			break;
		case TASK_INTERRUPTIBLE:
			stats->nr_sleeping++;
			break;
		case TASK_UNINTERRUPTIBLE:
			stats->nr_uninterruptible++;
			break;
		case TASK_STOPPED:
			stats->nr_stopped++;
			break;
		default:
			if (delayacct_is_task_waiting_on_io(tsk))
				stats->nr_io_wait++;
			break;
		}
	}
	cgroup_task_iter_end(cgrp, &it);

err:
	return ret;
}


/*
 * seq_file methods for the tasks/procs files. The seq_file position is the
 * next pid to display; the seq_file iterator is a pointer to the pid
 * in the cgroup->l->list array.
 */

static void *cgroup_pidlist_start(struct seq_file *s, loff_t *pos)
{
	/*
	 * Initially we receive a position value that corresponds to
	 * one more than the last pid shown (or 0 on the first call or
	 * after a seek to the start). Use a binary-search to find the
	 * next pid to display, if any
	 */
	struct cgroup_pidlist *l = s->private;
	int index = 0, pid = *pos;
	int *iter;

	down_read(&l->rwsem);
	if (pid) {
		int end = l->length;

		while (index < end) {
			int mid = (index + end) / 2;
			if (l->list[mid] == pid) {
				index = mid;
				break;
			} else if (l->list[mid] <= pid)
				index = mid + 1;
			else
				end = mid;
		}
	}
	/* If we're off the end of the array, we're done */
	if (index >= l->length)
		return NULL;
	/* Update the abstract position to be the actual pid that we found */
	iter = l->list + index;
	*pos = *iter;
	return iter;
}

static void cgroup_pidlist_stop(struct seq_file *s, void *v)
{
	struct cgroup_pidlist *l = s->private;
	up_read(&l->rwsem);
}

static void *cgroup_pidlist_next(struct seq_file *s, void *v, loff_t *pos)
{
	struct cgroup_pidlist *l = s->private;
	pid_t *p = v;
	pid_t *end = l->list + l->length;
	/*
	 * Advance to the next pid in the array. If this goes off the
	 * end, we're done
	 */
	p++;
	if (p >= end) {
		return NULL;
	} else {
		*pos = *p;
		return p;
	}
}

static int cgroup_pidlist_show(struct seq_file *s, void *v)
{
	return seq_printf(s, "%d\n", *(int *)v);
}

/*
 * seq_operations functions for iterating on pidlists through seq_file -
 * independent of whether it's tasks or procs
 */
static const struct seq_operations cgroup_pidlist_seq_operations = {
	.start = cgroup_pidlist_start,
	.stop = cgroup_pidlist_stop,
	.next = cgroup_pidlist_next,
	.show = cgroup_pidlist_show,
};

static void cgroup_release_pid_array(struct cgroup_pidlist *l)
{
	/*
	 * the case where we're the last user of this particular pidlist will
	 * have us remove it from the cgroup's list, which entails taking the
	 * mutex. since in pidlist_find the pidlist->lock depends on cgroup->
	 * pidlist_mutex, we have to take pidlist_mutex first.
	 */
	mutex_lock(&l->owner->pidlist_mutex);
	down_write(&l->rwsem);
	BUG_ON(!l->use_count);
	if (!--l->use_count) {
		/* we're the last user if refcount is 0; remove and free */
		list_del(&l->links);
		mutex_unlock(&l->owner->pidlist_mutex);
		pidlist_free(l->list);
		put_pid_ns(l->key.ns);
		up_write(&l->rwsem);
		kfree(l);
		return;
	}
	mutex_unlock(&l->owner->pidlist_mutex);
	up_write(&l->rwsem);
}

static int cgroup_pidlist_release(struct inode *inode, struct file *file)
{
	struct cgroup_pidlist *l;
	if (!(file->f_mode & FMODE_READ))
		return 0;
	/*
	 * the seq_file will only be initialized if the file was opened for
	 * reading; hence we check if it's not null only in that case.
	 */
	l = ((struct seq_file *)file->private_data)->private;
	cgroup_release_pid_array(l);
	return seq_release(inode, file);
}

static const struct file_operations cgroup_pidlist_operations = {
	.read = seq_read,
	.llseek = seq_lseek,
	.write = cgroup_file_write,
	.release = cgroup_pidlist_release,
};

/*
 * The following functions handle opens on a file that displays a pidlist
 * (tasks or procs). Prepare an array of the process/thread IDs of whoever's
 * in the cgroup.
 */
/* helper function for the two below it */
static int cgroup_pidlist_open(struct file *file, enum cgroup_filetype type)
{
	struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent);
	struct cgroup_pidlist *l;
	int retval;

	/* Nothing to do for write-only files */
	if (!(file->f_mode & FMODE_READ))
		return 0;

	/* have the array populated */
	retval = pidlist_array_load(cgrp, type, &l);
	if (retval)
		return retval;
	/* configure file information */
	file->f_op = &cgroup_pidlist_operations;

	retval = seq_open(file, &cgroup_pidlist_seq_operations);
	if (retval) {
		cgroup_release_pid_array(l);
		return retval;
	}
	((struct seq_file *)file->private_data)->private = l;
	return 0;
}
static int cgroup_tasks_open(struct inode *unused, struct file *file)
{
	return cgroup_pidlist_open(file, CGROUP_FILE_TASKS);
}
static int cgroup_procs_open(struct inode *unused, struct file *file)
{
	return cgroup_pidlist_open(file, CGROUP_FILE_PROCS);
}

static u64 cgroup_read_notify_on_release(struct cgroup_subsys_state *css,
					 struct cftype *cft)
{
	return notify_on_release(css->cgroup);
}

static int cgroup_write_notify_on_release(struct cgroup_subsys_state *css,
					  struct cftype *cft, u64 val)
{
	clear_bit(CGRP_RELEASABLE, &css->cgroup->flags);
	if (val)
		set_bit(CGRP_NOTIFY_ON_RELEASE, &css->cgroup->flags);
	else
		clear_bit(CGRP_NOTIFY_ON_RELEASE, &css->cgroup->flags);
	return 0;
}

/*
 * When dput() is called asynchronously, if umount has been done and
 * then deactivate_super() in cgroup_free_fn() kills the superblock,
 * there's a small window that vfs will see the root dentry with non-zero
 * refcnt and trigger BUG().
 *
 * That's why we hold a reference before dput() and drop it right after.
 */
static void cgroup_dput(struct cgroup *cgrp)
{
	struct super_block *sb = cgrp->root->sb;

	atomic_inc(&sb->s_active);
	dput(cgrp->dentry);
	deactivate_super(sb);
}

/*
 * Unregister event and free resources.
 *
 * Gets called from workqueue.
 */
static void cgroup_event_remove(struct work_struct *work)
{
	struct cgroup_event *event = container_of(work, struct cgroup_event,
			remove);
	struct cgroup *cgrp = event->cgrp;

	remove_wait_queue(event->wqh, &event->wait);

	event->cft->unregister_event(cgrp, event->cft, event->eventfd);

	/* Notify userspace the event is going away. */
	eventfd_signal(event->eventfd, 1);

	eventfd_ctx_put(event->eventfd);
	kfree(event);
	cgroup_dput(cgrp);
}

/*
 * Gets called on POLLHUP on eventfd when user closes it.
 *
 * Called with wqh->lock held and interrupts disabled.
 */
static int cgroup_event_wake(wait_queue_t *wait, unsigned mode,
		int sync, void *key)
{
	struct cgroup_event *event = container_of(wait,
			struct cgroup_event, wait);
	struct cgroup *cgrp = event->cgrp;
	unsigned long flags = (unsigned long)key;

	if (flags & POLLHUP) {
		/*
		 * If the event has been detached at cgroup removal, we
		 * can simply return knowing the other side will cleanup
		 * for us.
		 *
		 * We can't race against event freeing since the other
		 * side will require wqh->lock via remove_wait_queue(),
		 * which we hold.