[mirror_ubuntu-bionic-kernel.git] / kernel / cgroup / cgroup-internal.h

/* SPDX-License-Identifier: GPL-2.0 */
#ifndef __CGROUP_INTERNAL_H
#define __CGROUP_INTERNAL_H

#include <linux/cgroup.h>
#include <linux/kernfs.h>
#include <linux/workqueue.h>
#include <linux/list.h>
#include <linux/refcount.h>

/*
 * A cgroup can be associated with multiple css_sets as different tasks may
 * belong to different cgroups on different hierarchies.  In the other
 * direction, a css_set is naturally associated with multiple cgroups.
 * This M:N relationship is represented by the following link structure
 * which exists for each association and allows traversing the associations
 * from both sides.
 */
struct cgrp_cset_link {
	/* the cgroup and css_set this link associates */
	struct cgroup		*cgrp;
	struct css_set		*cset;

	/* list of cgrp_cset_links anchored at cgrp->cset_links */
	struct list_head	cset_link;

	/* list of cgrp_cset_links anchored at css_set->cgrp_links */
	struct list_head	cgrp_link;
};

/* used to track tasks and csets during migration */
struct cgroup_taskset {
	/* the src and dst cset list running through cset->mg_node */
	struct list_head	src_csets;
	struct list_head	dst_csets;

	/* the number of tasks in the set */
	int			nr_tasks;

	/* the subsys currently being processed */
	int			ssid;

	/*
	 * Fields for cgroup_taskset_*() iteration.
	 *
	 * Before migration is committed, the target migration tasks are on
	 * ->mg_tasks of the csets on ->src_csets.  After, on ->mg_tasks of
	 * the csets on ->dst_csets.  ->csets point to either ->src_csets
	 * or ->dst_csets depending on whether migration is committed.
	 *
	 * ->cur_csets and ->cur_task point to the current task position
	 * during iteration.
	 */
	struct list_head	*csets;
	struct css_set		*cur_cset;
	struct task_struct	*cur_task;
};

/* migration context also tracks preloading */
struct cgroup_mgctx {
	/*
	 * Preloaded source and destination csets.  Used to guarantee
	 * atomic success or failure on actual migration.
	 */
	struct list_head	preloaded_src_csets;
	struct list_head	preloaded_dst_csets;

	/* tasks and csets to migrate */
	struct cgroup_taskset	tset;

	/* subsystems affected by migration */
	u16			ss_mask;
};

#define CGROUP_TASKSET_INIT(tset)						\
{										\
	.src_csets		= LIST_HEAD_INIT(tset.src_csets),		\
	.dst_csets		= LIST_HEAD_INIT(tset.dst_csets),		\
	.csets			= &tset.src_csets,				\
}

#define CGROUP_MGCTX_INIT(name)							\
{										\
	LIST_HEAD_INIT(name.preloaded_src_csets),				\
	LIST_HEAD_INIT(name.preloaded_dst_csets),				\
	CGROUP_TASKSET_INIT(name.tset),						\
}

#define DEFINE_CGROUP_MGCTX(name)						\
	struct cgroup_mgctx name = CGROUP_MGCTX_INIT(name)

struct cgroup_sb_opts {
	u16 subsys_mask;
	unsigned int flags;
	char *release_agent;
	bool cpuset_clone_children;
	char *name;
	/* User explicitly requested empty subsystem */
	bool none;
};

extern struct mutex cgroup_mutex;
extern spinlock_t css_set_lock;
extern struct cgroup_subsys *cgroup_subsys[];
extern struct list_head cgroup_roots;
extern struct file_system_type cgroup_fs_type;

/* iterate across the hierarchies */
#define for_each_root(root)						\
	list_for_each_entry((root), &cgroup_roots, root_list)

/**
 * for_each_subsys - iterate all enabled cgroup subsystems
 * @ss: the iteration cursor
 * @ssid: the index of @ss, CGROUP_SUBSYS_COUNT after reaching the end
 */
#define for_each_subsys(ss, ssid)					\
	for ((ssid) = 0; (ssid) < CGROUP_SUBSYS_COUNT &&		\
	     (((ss) = cgroup_subsys[ssid]) || true); (ssid)++)

static inline bool cgroup_is_dead(const struct cgroup *cgrp)
{
	return !(cgrp->self.flags & CSS_ONLINE);
}

static inline bool notify_on_release(const struct cgroup *cgrp)
{
	return test_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
}

void put_css_set_locked(struct css_set *cset);

static inline void put_css_set(struct css_set *cset)
{
	unsigned long flags;

	/*
	 * Ensure that the refcount doesn't hit zero while any readers
	 * can see it. Similar to atomic_dec_and_lock(), but for an
	 * rwlock
	 */
	if (refcount_dec_not_one(&cset->refcount))
		return;

	spin_lock_irqsave(&css_set_lock, flags);
	put_css_set_locked(cset);
	spin_unlock_irqrestore(&css_set_lock, flags);
}

/*
 * refcounted get/put for css_set objects
 */
static inline void get_css_set(struct css_set *cset)
{
	refcount_inc(&cset->refcount);
}

bool cgroup_ssid_enabled(int ssid);
bool cgroup_on_dfl(const struct cgroup *cgrp);
bool cgroup_is_thread_root(struct cgroup *cgrp);
bool cgroup_is_threaded(struct cgroup *cgrp);

struct cgroup_root *cgroup_root_from_kf(struct kernfs_root *kf_root);
struct cgroup *task_cgroup_from_root(struct task_struct *task,
				     struct cgroup_root *root);
struct cgroup *cgroup_kn_lock_live(struct kernfs_node *kn, bool drain_offline);
void cgroup_kn_unlock(struct kernfs_node *kn);
int cgroup_path_ns_locked(struct cgroup *cgrp, char *buf, size_t buflen,
			  struct cgroup_namespace *ns);

void cgroup_free_root(struct cgroup_root *root);
void init_cgroup_root(struct cgroup_root *root, struct cgroup_sb_opts *opts);
int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask, int ref_flags);
int rebind_subsystems(struct cgroup_root *dst_root, u16 ss_mask);
struct dentry *cgroup_do_mount(struct file_system_type *fs_type, int flags,
			       struct cgroup_root *root, unsigned long magic,
			       struct cgroup_namespace *ns);

int cgroup_migrate_vet_dst(struct cgroup *dst_cgrp);
void cgroup_migrate_finish(struct cgroup_mgctx *mgctx);
void cgroup_migrate_add_src(struct css_set *src_cset, struct cgroup *dst_cgrp,
			    struct cgroup_mgctx *mgctx);
int cgroup_migrate_prepare_dst(struct cgroup_mgctx *mgctx);
int cgroup_migrate(struct task_struct *leader, bool threadgroup,
		   struct cgroup_mgctx *mgctx);

int cgroup_attach_task(struct cgroup *dst_cgrp, struct task_struct *leader,
		       bool threadgroup);
struct task_struct *cgroup_procs_write_start(char *buf, bool threadgroup)
	__acquires(&cgroup_threadgroup_rwsem);
void cgroup_procs_write_finish(struct task_struct *task)
	__releases(&cgroup_threadgroup_rwsem);

void cgroup_lock_and_drain_offline(struct cgroup *cgrp);

int cgroup_mkdir(struct kernfs_node *parent_kn, const char *name, umode_t mode);
int cgroup_rmdir(struct kernfs_node *kn);
int cgroup_show_path(struct seq_file *sf, struct kernfs_node *kf_node,
		     struct kernfs_root *kf_root);

int cgroup_task_count(const struct cgroup *cgrp);

/*
 * stat.c
 */
void cgroup_stat_flush(struct cgroup *cgrp);
int cgroup_stat_init(struct cgroup *cgrp);
void cgroup_stat_exit(struct cgroup *cgrp);
void cgroup_stat_show_cputime(struct seq_file *seq);
void cgroup_stat_boot(void);

/*
 * namespace.c
 */
extern const struct proc_ns_operations cgroupns_operations;

/*
 * cgroup-v1.c
 */
extern struct cftype cgroup1_base_files[];
extern const struct file_operations proc_cgroupstats_operations;
extern struct kernfs_syscall_ops cgroup1_kf_syscall_ops;

bool cgroup1_ssid_disabled(int ssid);
void cgroup1_pidlist_destroy_all(struct cgroup *cgrp);
void cgroup1_release_agent(struct work_struct *work);
void cgroup1_check_for_release(struct cgroup *cgrp);
struct dentry *cgroup1_mount(struct file_system_type *fs_type, int flags,
			     void *data, unsigned long magic,
			     struct cgroup_namespace *ns);

#endif /* __CGROUP_INTERNAL_H */
Commit	Line	Data
b2441318	1	/* SPDX-License-Identifier: GPL-2.0 */
0a268dbd TH	2	#ifndef __CGROUP_INTERNAL_H
	3	#define __CGROUP_INTERNAL_H
	4
	5	#include <linux/cgroup.h>
	6	#include <linux/kernfs.h>
	7	#include <linux/workqueue.h>
	8	#include <linux/list.h>
4b9502e6	9	#include <linux/refcount.h>
0a268dbd TH	10
	11	/*
	12	* A cgroup can be associated with multiple css_sets as different tasks may
	13	* belong to different cgroups on different hierarchies. In the other
	14	* direction, a css_set is naturally associated with multiple cgroups.
	15	* This M:N relationship is represented by the following link structure
	16	* which exists for each association and allows traversing the associations
	17	* from both sides.
	18	*/
	19	struct cgrp_cset_link {
	20	/* the cgroup and css_set this link associates */
	21	struct cgroup *cgrp;
	22	struct css_set *cset;
	23
	24	/* list of cgrp_cset_links anchored at cgrp->cset_links */
	25	struct list_head cset_link;
	26
	27	/* list of cgrp_cset_links anchored at css_set->cgrp_links */
	28	struct list_head cgrp_link;
	29	};
	30
e595cd70 TH	31	/* used to track tasks and csets during migration */
	32	struct cgroup_taskset {
	33	/* the src and dst cset list running through cset->mg_node */
	34	struct list_head src_csets;
	35	struct list_head dst_csets;
	36
61046727 TH	37	/* the number of tasks in the set */
	38	int nr_tasks;
	39
e595cd70 TH	40	/* the subsys currently being processed */
	41	int ssid;
	42
	43	/*
	44	* Fields for cgroup_taskset_*() iteration.
	45	*
	46	* Before migration is committed, the target migration tasks are on
	47	* ->mg_tasks of the csets on ->src_csets. After, on ->mg_tasks of
	48	* the csets on ->dst_csets. ->csets point to either ->src_csets
	49	* or ->dst_csets depending on whether migration is committed.
	50	*
	51	* ->cur_csets and ->cur_task point to the current task position
	52	* during iteration.
	53	*/
	54	struct list_head *csets;
	55	struct css_set *cur_cset;
	56	struct task_struct *cur_task;
	57	};
	58
	59	/* migration context also tracks preloading */
	60	struct cgroup_mgctx {
	61	/*
	62	* Preloaded source and destination csets. Used to guarantee
	63	* atomic success or failure on actual migration.
	64	*/
	65	struct list_head preloaded_src_csets;
	66	struct list_head preloaded_dst_csets;
	67
	68	/* tasks and csets to migrate */
	69	struct cgroup_taskset tset;
bfc2cf6f TH	70
	71	/* subsystems affected by migration */
	72	u16 ss_mask;
e595cd70 TH	73	};
	74
	75	#define CGROUP_TASKSET_INIT(tset) \
	76	{ \
	77	.src_csets = LIST_HEAD_INIT(tset.src_csets), \
	78	.dst_csets = LIST_HEAD_INIT(tset.dst_csets), \
	79	.csets = &tset.src_csets, \
	80	}
	81
	82	#define CGROUP_MGCTX_INIT(name) \
	83	{ \
	84	LIST_HEAD_INIT(name.preloaded_src_csets), \
	85	LIST_HEAD_INIT(name.preloaded_dst_csets), \
	86	CGROUP_TASKSET_INIT(name.tset), \
	87	}
	88
	89	#define DEFINE_CGROUP_MGCTX(name) \
	90	struct cgroup_mgctx name = CGROUP_MGCTX_INIT(name)
	91
1592c9b2 TH	92	struct cgroup_sb_opts {
	93	u16 subsys_mask;
	94	unsigned int flags;
	95	char *release_agent;
	96	bool cpuset_clone_children;
	97	char *name;
	98	/* User explicitly requested empty subsystem */
	99	bool none;
	100	};
	101
0a268dbd TH	102	extern struct mutex cgroup_mutex;
	103	extern spinlock_t css_set_lock;
	104	extern struct cgroup_subsys *cgroup_subsys[];
	105	extern struct list_head cgroup_roots;
	106	extern struct file_system_type cgroup_fs_type;
	107
	108	/* iterate across the hierarchies */
	109	#define for_each_root(root) \
	110	list_for_each_entry((root), &cgroup_roots, root_list)
	111
	112	/**
	113	* for_each_subsys - iterate all enabled cgroup subsystems
	114	* @ss: the iteration cursor
	115	* @ssid: the index of @ss, CGROUP_SUBSYS_COUNT after reaching the end
	116	*/
	117	#define for_each_subsys(ss, ssid) \
	118	for ((ssid) = 0; (ssid) < CGROUP_SUBSYS_COUNT && \
	119	(((ss) = cgroup_subsys[ssid]) \|\| true); (ssid)++)
	120
	121	static inline bool cgroup_is_dead(const struct cgroup *cgrp)
	122	{
	123	return !(cgrp->self.flags & CSS_ONLINE);
	124	}
	125
	126	static inline bool notify_on_release(const struct cgroup *cgrp)
	127	{
	128	return test_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
	129	}
	130
dcfe149b TH	131	void put_css_set_locked(struct css_set *cset);
	132
	133	static inline void put_css_set(struct css_set *cset)
	134	{
	135	unsigned long flags;
	136
	137	/*
	138	* Ensure that the refcount doesn't hit zero while any readers
	139	* can see it. Similar to atomic_dec_and_lock(), but for an
	140	* rwlock
	141	*/
4b9502e6	142	if (refcount_dec_not_one(&cset->refcount))
dcfe149b TH	143	return;
	144
	145	spin_lock_irqsave(&css_set_lock, flags);
	146	put_css_set_locked(cset);
	147	spin_unlock_irqrestore(&css_set_lock, flags);
	148	}
	149
	150	/*
	151	* refcounted get/put for css_set objects
	152	*/
	153	static inline void get_css_set(struct css_set *cset)
	154	{
4b9502e6	155	refcount_inc(&cset->refcount);
dcfe149b TH	156	}
dcfe149b TH	157
0a268dbd TH	158	bool cgroup_ssid_enabled(int ssid);
0a268dbd TH	159	bool cgroup_on_dfl(const struct cgroup *cgrp);
7a0cf0e7 WL	160	bool cgroup_is_thread_root(struct cgroup *cgrp);
7a0cf0e7 WL	161	bool cgroup_is_threaded(struct cgroup *cgrp);
0a268dbd TH	162
	163	struct cgroup_root cgroup_root_from_kf(struct kernfs_root kf_root);
	164	struct cgroup task_cgroup_from_root(struct task_struct task,
	165	struct cgroup_root *root);
	166	struct cgroup cgroup_kn_lock_live(struct kernfs_node kn, bool drain_offline);
	167	void cgroup_kn_unlock(struct kernfs_node *kn);
	168	int cgroup_path_ns_locked(struct cgroup cgrp, char buf, size_t buflen,
	169	struct cgroup_namespace *ns);
	170
1592c9b2 TH	171	void cgroup_free_root(struct cgroup_root *root);
1592c9b2 TH	172	void init_cgroup_root(struct cgroup_root root, struct cgroup_sb_opts opts);
9732adc5	173	int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask, int ref_flags);
0a268dbd	174	int rebind_subsystems(struct cgroup_root *dst_root, u16 ss_mask);
1592c9b2 TH	175	struct dentry cgroup_do_mount(struct file_system_type fs_type, int flags,
	176	struct cgroup_root *root, unsigned long magic,
	177	struct cgroup_namespace *ns);
0a268dbd	178
8cfd8147	179	int cgroup_migrate_vet_dst(struct cgroup *dst_cgrp);
e595cd70 TH	180	void cgroup_migrate_finish(struct cgroup_mgctx *mgctx);
	181	void cgroup_migrate_add_src(struct css_set src_cset, struct cgroup dst_cgrp,
	182	struct cgroup_mgctx *mgctx);
	183	int cgroup_migrate_prepare_dst(struct cgroup_mgctx *mgctx);
0a268dbd	184	int cgroup_migrate(struct task_struct *leader, bool threadgroup,
bfc2cf6f	185	struct cgroup_mgctx *mgctx);
0a268dbd TH	186
	187	int cgroup_attach_task(struct cgroup dst_cgrp, struct task_struct leader,
	188	bool threadgroup);
715c809d TH	189	struct task_struct cgroup_procs_write_start(char buf, bool threadgroup)
	190	__acquires(&cgroup_threadgroup_rwsem);
	191	void cgroup_procs_write_finish(struct task_struct *task)
	192	__releases(&cgroup_threadgroup_rwsem);
0a268dbd TH	193
	194	void cgroup_lock_and_drain_offline(struct cgroup *cgrp);
	195
1592c9b2 TH	196	int cgroup_mkdir(struct kernfs_node parent_kn, const char name, umode_t mode);
	197	int cgroup_rmdir(struct kernfs_node *kn);
	198	int cgroup_show_path(struct seq_file sf, struct kernfs_node kf_node,
	199	struct kernfs_root *kf_root);
	200
a28f8f5e WL	201	int cgroup_task_count(const struct cgroup *cgrp);
a28f8f5e WL	202
041cd640 TH	203	/*
	204	* stat.c
	205	*/
	206	void cgroup_stat_flush(struct cgroup *cgrp);
	207	int cgroup_stat_init(struct cgroup *cgrp);
	208	void cgroup_stat_exit(struct cgroup *cgrp);
d41bf8c9	209	void cgroup_stat_show_cputime(struct seq_file *seq);
041cd640 TH	210	void cgroup_stat_boot(void);
041cd640 TH	211
dcfe149b TH	212	/*
	213	* namespace.c
	214	*/
	215	extern const struct proc_ns_operations cgroupns_operations;
	216
0a268dbd TH	217	/*
	218	* cgroup-v1.c
	219	*/
d62beb7f	220	extern struct cftype cgroup1_base_files[];
0a268dbd	221	extern const struct file_operations proc_cgroupstats_operations;
1592c9b2	222	extern struct kernfs_syscall_ops cgroup1_kf_syscall_ops;
0a268dbd	223
d62beb7f TH	224	bool cgroup1_ssid_disabled(int ssid);
	225	void cgroup1_pidlist_destroy_all(struct cgroup *cgrp);
	226	void cgroup1_release_agent(struct work_struct *work);
	227	void cgroup1_check_for_release(struct cgroup *cgrp);
1592c9b2 TH	228	struct dentry cgroup1_mount(struct file_system_type fs_type, int flags,
	229	void *data, unsigned long magic,
	230	struct cgroup_namespace *ns);
0a268dbd TH	231
0a268dbd TH	232	#endif /* __CGROUP_INTERNAL_H */