---------------------------
-What: USER_SCHED
-When: 2.6.34
-
-Why: USER_SCHED was implemented as a proof of concept for group scheduling.
- The effect of USER_SCHED can already be achieved from userspace with
- the help of libcgroup. The removal of USER_SCHED will also simplify
- the scheduler code with the removal of one major ifdef. There are also
- issues USER_SCHED has with USER_NS. A decision was taken not to fix
- those and instead remove USER_SCHED. Also new group scheduling
- features will not be implemented for USER_SCHED.
-
-Who: Dhaval Giani <dhaval@linux.vnet.ibm.com>
-
----------------------------
-
What: PRISM54
When: 2.6.34
uid_t uid;
struct user_namespace *user_ns;
-#ifdef CONFIG_USER_SCHED
- struct task_group *tg;
-#ifdef CONFIG_SYSFS
- struct kobject kobj;
- struct delayed_work work;
-#endif
-#endif
-
#ifdef CONFIG_PERF_EVENTS
atomic_long_t locked_vm;
#endif
extern void normalize_rt_tasks(void);
-#ifdef CONFIG_GROUP_SCHED
+#ifdef CONFIG_CGROUP_SCHED
extern struct task_group init_task_group;
-#ifdef CONFIG_USER_SCHED
-extern struct task_group root_task_group;
-extern void set_tg_uid(struct user_struct *user);
-#endif
extern struct task_group *sched_create_group(struct task_group *parent);
extern void sched_destroy_group(struct task_group *tg);
config HAVE_UNSTABLE_SCHED_CLOCK
bool
-config GROUP_SCHED
- bool "Group CPU scheduler"
- depends on EXPERIMENTAL
- default n
- help
- This feature lets CPU scheduler recognize task groups and control CPU
- bandwidth allocation to such task groups.
- In order to create a group from arbitrary set of processes, use
- CONFIG_CGROUPS. (See Control Group support.)
-
-config FAIR_GROUP_SCHED
- bool "Group scheduling for SCHED_OTHER"
- depends on GROUP_SCHED
- default GROUP_SCHED
-
-config RT_GROUP_SCHED
- bool "Group scheduling for SCHED_RR/FIFO"
- depends on EXPERIMENTAL
- depends on GROUP_SCHED
- default n
- help
- This feature lets you explicitly allocate real CPU bandwidth
- to users or control groups (depending on the "Basis for grouping tasks"
- setting below. If enabled, it will also make it impossible to
- schedule realtime tasks for non-root users until you allocate
- realtime bandwidth for them.
- See Documentation/scheduler/sched-rt-group.txt for more information.
-
-choice
- depends on GROUP_SCHED
- prompt "Basis for grouping tasks"
- default USER_SCHED
-
-config USER_SCHED
- bool "user id"
- help
- This option will choose userid as the basis for grouping
- tasks, thus providing equal CPU bandwidth to each user.
-
-config CGROUP_SCHED
- bool "Control groups"
- depends on CGROUPS
- help
- This option allows you to create arbitrary task groups
- using the "cgroup" pseudo filesystem and control
- the cpu bandwidth allocated to each such task group.
- Refer to Documentation/cgroups/cgroups.txt for more
- information on "cgroup" pseudo filesystem.
-
-endchoice
-
menuconfig CGROUPS
boolean "Control Group support"
help
Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
size is 4096bytes, 512k per 1Gbytes of swap.
+menuconfig CGROUP_SCHED
+ bool "Group CPU scheduler"
+ depends on EXPERIMENTAL && CGROUPS
+ default n
+ help
+ This feature lets CPU scheduler recognize task groups and control CPU
+ bandwidth allocation to such task groups. It uses cgroups to group
+ tasks.
+
+if CGROUP_SCHED
+config FAIR_GROUP_SCHED
+ bool "Group scheduling for SCHED_OTHER"
+ depends on CGROUP_SCHED
+ default CGROUP_SCHED
+
+config RT_GROUP_SCHED
+ bool "Group scheduling for SCHED_RR/FIFO"
+ depends on EXPERIMENTAL
+ depends on CGROUP_SCHED
+ default n
+ help
+ This feature lets you explicitly allocate real CPU bandwidth
+ to users or control groups (depending on the "Basis for grouping tasks"
+ setting below. If enabled, it will also make it impossible to
+ schedule realtime tasks for non-root users until you allocate
+ realtime bandwidth for them.
+ See Documentation/scheduler/sched-rt-group.txt for more information.
+
+endif #CGROUP_SCHED
+
endif # CGROUPS
config MM_OWNER
goto group_exit;
}
- /* create the /sys/kernel/uids/ directory */
- error = uids_sysfs_init();
- if (error)
- goto notes_exit;
-
return 0;
-notes_exit:
- if (notes_size > 0)
- sysfs_remove_bin_file(kernel_kobj, ¬es_attr);
group_exit:
sysfs_remove_group(kernel_kobj, &kernel_attr_group);
kset_exit:
*/
static DEFINE_MUTEX(sched_domains_mutex);
-#ifdef CONFIG_GROUP_SCHED
+#ifdef CONFIG_CGROUP_SCHED
#include <linux/cgroup.h>
/* task group related information */
struct task_group {
-#ifdef CONFIG_CGROUP_SCHED
struct cgroup_subsys_state css;
-#endif
-
-#ifdef CONFIG_USER_SCHED
- uid_t uid;
-#endif
#ifdef CONFIG_FAIR_GROUP_SCHED
/* schedulable entities of this group on each cpu */
struct list_head children;
};
-#ifdef CONFIG_USER_SCHED
-
-/* Helper function to pass uid information to create_sched_user() */
-void set_tg_uid(struct user_struct *user)
-{
- user->tg->uid = user->uid;
-}
-
-/*
- * Root task group.
- * Every UID task group (including init_task_group aka UID-0) will
- * be a child to this group.
- */
-struct task_group root_task_group;
-
-#ifdef CONFIG_FAIR_GROUP_SCHED
-/* Default task group's sched entity on each cpu */
-static DEFINE_PER_CPU(struct sched_entity, init_sched_entity);
-/* Default task group's cfs_rq on each cpu */
-static DEFINE_PER_CPU_SHARED_ALIGNED(struct cfs_rq, init_tg_cfs_rq);
-#endif /* CONFIG_FAIR_GROUP_SCHED */
-
-#ifdef CONFIG_RT_GROUP_SCHED
-static DEFINE_PER_CPU(struct sched_rt_entity, init_sched_rt_entity);
-static DEFINE_PER_CPU_SHARED_ALIGNED(struct rt_rq, init_rt_rq_var);
-#endif /* CONFIG_RT_GROUP_SCHED */
-#else /* !CONFIG_USER_SCHED */
#define root_task_group init_task_group
-#endif /* CONFIG_USER_SCHED */
/* task_group_lock serializes add/remove of task groups and also changes to
* a task group's cpu shares.
}
#endif
-#ifdef CONFIG_USER_SCHED
-# define INIT_TASK_GROUP_LOAD (2*NICE_0_LOAD)
-#else /* !CONFIG_USER_SCHED */
# define INIT_TASK_GROUP_LOAD NICE_0_LOAD
-#endif /* CONFIG_USER_SCHED */
/*
* A weight of 0 or 1 can cause arithmetics problems.
{
struct task_group *tg;
-#ifdef CONFIG_USER_SCHED
- rcu_read_lock();
- tg = __task_cred(p)->user->tg;
- rcu_read_unlock();
-#elif defined(CONFIG_CGROUP_SCHED)
+#ifdef CONFIG_CGROUP_SCHED
tg = container_of(task_subsys_state(p, cpu_cgroup_subsys_id),
struct task_group, css);
#else
return NULL;
}
-#endif /* CONFIG_GROUP_SCHED */
+#endif /* CONFIG_CGROUP_SCHED */
/* CFS-related fields in a runqueue */
struct cfs_rq {
#ifdef CONFIG_RT_GROUP_SCHED
alloc_size += 2 * nr_cpu_ids * sizeof(void **);
#endif
-#ifdef CONFIG_USER_SCHED
- alloc_size *= 2;
-#endif
#ifdef CONFIG_CPUMASK_OFFSTACK
alloc_size += num_possible_cpus() * cpumask_size();
#endif
init_task_group.cfs_rq = (struct cfs_rq **)ptr;
ptr += nr_cpu_ids * sizeof(void **);
-#ifdef CONFIG_USER_SCHED
- root_task_group.se = (struct sched_entity **)ptr;
- ptr += nr_cpu_ids * sizeof(void **);
-
- root_task_group.cfs_rq = (struct cfs_rq **)ptr;
- ptr += nr_cpu_ids * sizeof(void **);
-#endif /* CONFIG_USER_SCHED */
#endif /* CONFIG_FAIR_GROUP_SCHED */
#ifdef CONFIG_RT_GROUP_SCHED
init_task_group.rt_se = (struct sched_rt_entity **)ptr;
init_task_group.rt_rq = (struct rt_rq **)ptr;
ptr += nr_cpu_ids * sizeof(void **);
-#ifdef CONFIG_USER_SCHED
- root_task_group.rt_se = (struct sched_rt_entity **)ptr;
- ptr += nr_cpu_ids * sizeof(void **);
-
- root_task_group.rt_rq = (struct rt_rq **)ptr;
- ptr += nr_cpu_ids * sizeof(void **);
-#endif /* CONFIG_USER_SCHED */
#endif /* CONFIG_RT_GROUP_SCHED */
#ifdef CONFIG_CPUMASK_OFFSTACK
for_each_possible_cpu(i) {
#ifdef CONFIG_RT_GROUP_SCHED
init_rt_bandwidth(&init_task_group.rt_bandwidth,
global_rt_period(), global_rt_runtime());
-#ifdef CONFIG_USER_SCHED
- init_rt_bandwidth(&root_task_group.rt_bandwidth,
- global_rt_period(), RUNTIME_INF);
-#endif /* CONFIG_USER_SCHED */
#endif /* CONFIG_RT_GROUP_SCHED */
-#ifdef CONFIG_GROUP_SCHED
+#ifdef CONFIG_CGROUP_SCHED
list_add(&init_task_group.list, &task_groups);
INIT_LIST_HEAD(&init_task_group.children);
-#ifdef CONFIG_USER_SCHED
- INIT_LIST_HEAD(&root_task_group.children);
- init_task_group.parent = &root_task_group;
- list_add(&init_task_group.siblings, &root_task_group.children);
-#endif /* CONFIG_USER_SCHED */
-#endif /* CONFIG_GROUP_SCHED */
+#endif /* CONFIG_CGROUP_SCHED */
#if defined CONFIG_FAIR_GROUP_SCHED && defined CONFIG_SMP
update_shares_data = __alloc_percpu(nr_cpu_ids * sizeof(unsigned long),
* directly in rq->cfs (i.e init_task_group->se[] = NULL).
*/
init_tg_cfs_entry(&init_task_group, &rq->cfs, NULL, i, 1, NULL);
-#elif defined CONFIG_USER_SCHED
- root_task_group.shares = NICE_0_LOAD;
- init_tg_cfs_entry(&root_task_group, &rq->cfs, NULL, i, 0, NULL);
- /*
- * In case of task-groups formed thr' the user id of tasks,
- * init_task_group represents tasks belonging to root user.
- * Hence it forms a sibling of all subsequent groups formed.
- * In this case, init_task_group gets only a fraction of overall
- * system cpu resource, based on the weight assigned to root
- * user's cpu share (INIT_TASK_GROUP_LOAD). This is accomplished
- * by letting tasks of init_task_group sit in a separate cfs_rq
- * (init_tg_cfs_rq) and having one entity represent this group of
- * tasks in rq->cfs (i.e init_task_group->se[] != NULL).
- */
- init_tg_cfs_entry(&init_task_group,
- &per_cpu(init_tg_cfs_rq, i),
- &per_cpu(init_sched_entity, i), i, 1,
- root_task_group.se[i]);
-
#endif
#endif /* CONFIG_FAIR_GROUP_SCHED */
INIT_LIST_HEAD(&rq->leaf_rt_rq_list);
#ifdef CONFIG_CGROUP_SCHED
init_tg_rt_entry(&init_task_group, &rq->rt, NULL, i, 1, NULL);
-#elif defined CONFIG_USER_SCHED
- init_tg_rt_entry(&root_task_group, &rq->rt, NULL, i, 0, NULL);
- init_tg_rt_entry(&init_task_group,
- &per_cpu(init_rt_rq_var, i),
- &per_cpu(init_sched_rt_entity, i), i, 1,
- root_task_group.rt_se[i]);
#endif
#endif
}
#endif /* CONFIG_RT_GROUP_SCHED */
-#ifdef CONFIG_GROUP_SCHED
+#ifdef CONFIG_CGROUP_SCHED
static void free_sched_group(struct task_group *tg)
{
free_fair_sched_group(tg);
task_rq_unlock(rq, &flags);
}
-#endif /* CONFIG_GROUP_SCHED */
+#endif /* CONFIG_CGROUP_SCHED */
#ifdef CONFIG_FAIR_GROUP_SCHED
static void __set_se_shares(struct sched_entity *se, unsigned long shares)
runtime = d->rt_runtime;
}
-#ifdef CONFIG_USER_SCHED
- if (tg == &root_task_group) {
- period = global_rt_period();
- runtime = global_rt_runtime();
- }
-#endif
-
/*
* Cannot have more runtime than the period.
*/
if (!new_user)
return -EAGAIN;
- if (!task_can_switch_user(new_user, current)) {
- free_uid(new_user);
- return -EINVAL;
- }
-
if (atomic_read(&new_user->processes) >=
current->signal->rlim[RLIMIT_NPROC].rlim_cur &&
new_user != INIT_USER) {
.sigpending = ATOMIC_INIT(0),
.locked_shm = 0,
.user_ns = &init_user_ns,
-#ifdef CONFIG_USER_SCHED
- .tg = &init_task_group,
-#endif
};
/*
put_user_ns(up->user_ns);
}
-#ifdef CONFIG_USER_SCHED
-
-static void sched_destroy_user(struct user_struct *up)
-{
- sched_destroy_group(up->tg);
-}
-
-static int sched_create_user(struct user_struct *up)
-{
- int rc = 0;
-
- up->tg = sched_create_group(&root_task_group);
- if (IS_ERR(up->tg))
- rc = -ENOMEM;
-
- set_tg_uid(up);
-
- return rc;
-}
-
-#else /* CONFIG_USER_SCHED */
-
-static void sched_destroy_user(struct user_struct *up) { }
-static int sched_create_user(struct user_struct *up) { return 0; }
-
-#endif /* CONFIG_USER_SCHED */
-
-#if defined(CONFIG_USER_SCHED) && defined(CONFIG_SYSFS)
-
-static struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent)
-{
- struct user_struct *user;
- struct hlist_node *h;
-
- hlist_for_each_entry(user, h, hashent, uidhash_node) {
- if (user->uid == uid) {
- /* possibly resurrect an "almost deleted" object */
- if (atomic_inc_return(&user->__count) == 1)
- cancel_delayed_work(&user->work);
- return user;
- }
- }
-
- return NULL;
-}
-
-static struct kset *uids_kset; /* represents the /sys/kernel/uids/ directory */
-static DEFINE_MUTEX(uids_mutex);
-
-static inline void uids_mutex_lock(void)
-{
- mutex_lock(&uids_mutex);
-}
-
-static inline void uids_mutex_unlock(void)
-{
- mutex_unlock(&uids_mutex);
-}
-
-/* uid directory attributes */
-#ifdef CONFIG_FAIR_GROUP_SCHED
-static ssize_t cpu_shares_show(struct kobject *kobj,
- struct kobj_attribute *attr,
- char *buf)
-{
- struct user_struct *up = container_of(kobj, struct user_struct, kobj);
-
- return sprintf(buf, "%lu\n", sched_group_shares(up->tg));
-}
-
-static ssize_t cpu_shares_store(struct kobject *kobj,
- struct kobj_attribute *attr,
- const char *buf, size_t size)
-{
- struct user_struct *up = container_of(kobj, struct user_struct, kobj);
- unsigned long shares;
- int rc;
-
- sscanf(buf, "%lu", &shares);
-
- rc = sched_group_set_shares(up->tg, shares);
-
- return (rc ? rc : size);
-}
-
-static struct kobj_attribute cpu_share_attr =
- __ATTR(cpu_share, 0644, cpu_shares_show, cpu_shares_store);
-#endif
-
-#ifdef CONFIG_RT_GROUP_SCHED
-static ssize_t cpu_rt_runtime_show(struct kobject *kobj,
- struct kobj_attribute *attr,
- char *buf)
-{
- struct user_struct *up = container_of(kobj, struct user_struct, kobj);
-
- return sprintf(buf, "%ld\n", sched_group_rt_runtime(up->tg));
-}
-
-static ssize_t cpu_rt_runtime_store(struct kobject *kobj,
- struct kobj_attribute *attr,
- const char *buf, size_t size)
-{
- struct user_struct *up = container_of(kobj, struct user_struct, kobj);
- unsigned long rt_runtime;
- int rc;
-
- sscanf(buf, "%ld", &rt_runtime);
-
- rc = sched_group_set_rt_runtime(up->tg, rt_runtime);
-
- return (rc ? rc : size);
-}
-
-static struct kobj_attribute cpu_rt_runtime_attr =
- __ATTR(cpu_rt_runtime, 0644, cpu_rt_runtime_show, cpu_rt_runtime_store);
-
-static ssize_t cpu_rt_period_show(struct kobject *kobj,
- struct kobj_attribute *attr,
- char *buf)
-{
- struct user_struct *up = container_of(kobj, struct user_struct, kobj);
-
- return sprintf(buf, "%lu\n", sched_group_rt_period(up->tg));
-}
-
-static ssize_t cpu_rt_period_store(struct kobject *kobj,
- struct kobj_attribute *attr,
- const char *buf, size_t size)
-{
- struct user_struct *up = container_of(kobj, struct user_struct, kobj);
- unsigned long rt_period;
- int rc;
-
- sscanf(buf, "%lu", &rt_period);
-
- rc = sched_group_set_rt_period(up->tg, rt_period);
-
- return (rc ? rc : size);
-}
-
-static struct kobj_attribute cpu_rt_period_attr =
- __ATTR(cpu_rt_period, 0644, cpu_rt_period_show, cpu_rt_period_store);
-#endif
-
-/* default attributes per uid directory */
-static struct attribute *uids_attributes[] = {
-#ifdef CONFIG_FAIR_GROUP_SCHED
- &cpu_share_attr.attr,
-#endif
-#ifdef CONFIG_RT_GROUP_SCHED
- &cpu_rt_runtime_attr.attr,
- &cpu_rt_period_attr.attr,
-#endif
- NULL
-};
-
-/* the lifetime of user_struct is not managed by the core (now) */
-static void uids_release(struct kobject *kobj)
-{
- return;
-}
-
-static struct kobj_type uids_ktype = {
- .sysfs_ops = &kobj_sysfs_ops,
- .default_attrs = uids_attributes,
- .release = uids_release,
-};
-
-/*
- * Create /sys/kernel/uids/<uid>/cpu_share file for this user
- * We do not create this file for users in a user namespace (until
- * sysfs tagging is implemented).
- *
- * See Documentation/scheduler/sched-design-CFS.txt for ramifications.
- */
-static int uids_user_create(struct user_struct *up)
-{
- struct kobject *kobj = &up->kobj;
- int error;
-
- memset(kobj, 0, sizeof(struct kobject));
- if (up->user_ns != &init_user_ns)
- return 0;
- kobj->kset = uids_kset;
- error = kobject_init_and_add(kobj, &uids_ktype, NULL, "%d", up->uid);
- if (error) {
- kobject_put(kobj);
- goto done;
- }
-
- kobject_uevent(kobj, KOBJ_ADD);
-done:
- return error;
-}
-
-/* create these entries in sysfs:
- * "/sys/kernel/uids" directory
- * "/sys/kernel/uids/0" directory (for root user)
- * "/sys/kernel/uids/0/cpu_share" file (for root user)
- */
-int __init uids_sysfs_init(void)
-{
- uids_kset = kset_create_and_add("uids", NULL, kernel_kobj);
- if (!uids_kset)
- return -ENOMEM;
-
- return uids_user_create(&root_user);
-}
-
-/* delayed work function to remove sysfs directory for a user and free up
- * corresponding structures.
- */
-static void cleanup_user_struct(struct work_struct *w)
-{
- struct user_struct *up = container_of(w, struct user_struct, work.work);
- unsigned long flags;
- int remove_user = 0;
-
- /* Make uid_hash_remove() + sysfs_remove_file() + kobject_del()
- * atomic.
- */
- uids_mutex_lock();
-
- spin_lock_irqsave(&uidhash_lock, flags);
- if (atomic_read(&up->__count) == 0) {
- uid_hash_remove(up);
- remove_user = 1;
- }
- spin_unlock_irqrestore(&uidhash_lock, flags);
-
- if (!remove_user)
- goto done;
-
- if (up->user_ns == &init_user_ns) {
- kobject_uevent(&up->kobj, KOBJ_REMOVE);
- kobject_del(&up->kobj);
- kobject_put(&up->kobj);
- }
-
- sched_destroy_user(up);
- key_put(up->uid_keyring);
- key_put(up->session_keyring);
- kmem_cache_free(uid_cachep, up);
-
-done:
- uids_mutex_unlock();
-}
-
-/* IRQs are disabled and uidhash_lock is held upon function entry.
- * IRQ state (as stored in flags) is restored and uidhash_lock released
- * upon function exit.
- */
-static void free_user(struct user_struct *up, unsigned long flags)
-{
- INIT_DELAYED_WORK(&up->work, cleanup_user_struct);
- schedule_delayed_work(&up->work, msecs_to_jiffies(1000));
- spin_unlock_irqrestore(&uidhash_lock, flags);
-}
-
-#else /* CONFIG_USER_SCHED && CONFIG_SYSFS */
-
static struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent)
{
struct user_struct *user;
return NULL;
}
-int uids_sysfs_init(void) { return 0; }
-static inline int uids_user_create(struct user_struct *up) { return 0; }
-static inline void uids_mutex_lock(void) { }
-static inline void uids_mutex_unlock(void) { }
-
/* IRQs are disabled and uidhash_lock is held upon function entry.
* IRQ state (as stored in flags) is restored and uidhash_lock released
* upon function exit.
{
uid_hash_remove(up);
spin_unlock_irqrestore(&uidhash_lock, flags);
- sched_destroy_user(up);
key_put(up->uid_keyring);
key_put(up->session_keyring);
kmem_cache_free(uid_cachep, up);
}
-#endif
-
-#if defined(CONFIG_RT_GROUP_SCHED) && defined(CONFIG_USER_SCHED)
-/*
- * We need to check if a setuid can take place. This function should be called
- * before successfully completing the setuid.
- */
-int task_can_switch_user(struct user_struct *up, struct task_struct *tsk)
-{
-
- return sched_rt_can_attach(up->tg, tsk);
-
-}
-#else
-int task_can_switch_user(struct user_struct *up, struct task_struct *tsk)
-{
- return 1;
-}
-#endif
-
/*
* Locate the user_struct for the passed UID. If found, take a ref on it. The
* caller must undo that ref with free_uid().
/* Make uid_hash_find() + uids_user_create() + uid_hash_insert()
* atomic.
*/
- uids_mutex_lock();
-
spin_lock_irq(&uidhash_lock);
up = uid_hash_find(uid, hashent);
spin_unlock_irq(&uidhash_lock);
new->uid = uid;
atomic_set(&new->__count, 1);
- if (sched_create_user(new) < 0)
- goto out_free_user;
-
new->user_ns = get_user_ns(ns);
- if (uids_user_create(new))
- goto out_destoy_sched;
-
/*
* Before adding this, check whether we raced
* on adding the same user already..
spin_unlock_irq(&uidhash_lock);
}
- uids_mutex_unlock();
-
return up;
-out_destoy_sched:
- sched_destroy_user(new);
put_user_ns(new->user_ns);
-out_free_user:
kmem_cache_free(uid_cachep, new);
out_unlock:
- uids_mutex_unlock();
return NULL;
}
projects / linux-2.6-block.git / commitdiff