autogroup.c 6.5 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0
2 3 4
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/utsname.h>
5 6
#include <linux/security.h>
#include <linux/export.h>
7

8 9
#include "sched.h"

10 11 12 13
unsigned int __read_mostly sysctl_sched_autogroup_enabled = 1;
static struct autogroup autogroup_default;
static atomic_t autogroup_seq_nr;

14
void __init autogroup_init(struct task_struct *init_task)
15
{
16
	autogroup_default.tg = &root_task_group;
17 18 19 20 21
	kref_init(&autogroup_default.kref);
	init_rwsem(&autogroup_default.lock);
	init_task->signal->autogroup = &autogroup_default;
}

22
void autogroup_free(struct task_group *tg)
23 24 25 26 27 28 29 30
{
	kfree(tg->autogroup);
}

static inline void autogroup_destroy(struct kref *kref)
{
	struct autogroup *ag = container_of(kref, struct autogroup, kref);

31 32 33 34 35
#ifdef CONFIG_RT_GROUP_SCHED
	/* We've redirected RT tasks to the root task group... */
	ag->tg->rt_se = NULL;
	ag->tg->rt_rq = NULL;
#endif
36
	sched_offline_group(ag->tg);
37 38 39 40 41 42 43 44 45 46 47 48 49 50
	sched_destroy_group(ag->tg);
}

static inline void autogroup_kref_put(struct autogroup *ag)
{
	kref_put(&ag->kref, autogroup_destroy);
}

static inline struct autogroup *autogroup_kref_get(struct autogroup *ag)
{
	kref_get(&ag->kref);
	return ag;
}

51 52 53 54 55 56 57 58 59 60 61 62 63 64
static inline struct autogroup *autogroup_task_get(struct task_struct *p)
{
	struct autogroup *ag;
	unsigned long flags;

	if (!lock_task_sighand(p, &flags))
		return autogroup_kref_get(&autogroup_default);

	ag = autogroup_kref_get(p->signal->autogroup);
	unlock_task_sighand(p, &flags);

	return ag;
}

65 66 67 68 69 70 71 72
static inline struct autogroup *autogroup_create(void)
{
	struct autogroup *ag = kzalloc(sizeof(*ag), GFP_KERNEL);
	struct task_group *tg;

	if (!ag)
		goto out_fail;

73
	tg = sched_create_group(&root_task_group);
74 75 76 77 78 79 80
	if (IS_ERR(tg))
		goto out_free;

	kref_init(&ag->kref);
	init_rwsem(&ag->lock);
	ag->id = atomic_inc_return(&autogroup_seq_nr);
	ag->tg = tg;
81 82 83 84 85 86
#ifdef CONFIG_RT_GROUP_SCHED
	/*
	 * Autogroup RT tasks are redirected to the root task group
	 * so we don't have to move tasks around upon policy change,
	 * or flail around trying to allocate bandwidth on the fly.
	 * A bandwidth exception in __sched_setscheduler() allows
87
	 * the policy change to proceed.
88 89 90 91 92
	 */
	free_rt_sched_group(tg);
	tg->rt_se = root_task_group.rt_se;
	tg->rt_rq = root_task_group.rt_rq;
#endif
93 94
	tg->autogroup = ag;

95
	sched_online_group(tg, &root_task_group);
96 97 98 99 100 101 102
	return ag;

out_free:
	kfree(ag);
out_fail:
	if (printk_ratelimit()) {
		printk(KERN_WARNING "autogroup_create: %s failure.\n",
103
			ag ? "sched_create_group()" : "kzalloc()");
104 105 106 107 108
	}

	return autogroup_kref_get(&autogroup_default);
}

109
bool task_wants_autogroup(struct task_struct *p, struct task_group *tg)
110 111 112 113
{
	if (tg != &root_task_group)
		return false;
	/*
114 115 116
	 * If we race with autogroup_move_group() the caller can use the old
	 * value of signal->autogroup but in this case sched_move_task() will
	 * be called again before autogroup_kref_put().
117 118 119
	 *
	 * However, there is no way sched_autogroup_exit_task() could tell us
	 * to avoid autogroup->tg, so we abuse PF_EXITING flag for this case.
120
	 */
121 122 123
	if (p->flags & PF_EXITING)
		return false;

124 125 126
	return true;
}

127 128 129 130 131 132 133 134 135 136
void sched_autogroup_exit_task(struct task_struct *p)
{
	/*
	 * We are going to call exit_notify() and autogroup_move_group() can't
	 * see this thread after that: we can no longer use signal->autogroup.
	 * See the PF_EXITING check in task_wants_autogroup().
	 */
	sched_move_task(p);
}

137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152
static void
autogroup_move_group(struct task_struct *p, struct autogroup *ag)
{
	struct autogroup *prev;
	struct task_struct *t;
	unsigned long flags;

	BUG_ON(!lock_task_sighand(p, &flags));

	prev = p->signal->autogroup;
	if (prev == ag) {
		unlock_task_sighand(p, &flags);
		return;
	}

	p->signal->autogroup = autogroup_kref_get(ag);
153 154 155 156 157 158 159
	/*
	 * We can't avoid sched_move_task() after we changed signal->autogroup,
	 * this process can already run with task_group() == prev->tg or we can
	 * race with cgroup code which can read autogroup = prev under rq->lock.
	 * In the latter case for_each_thread() can not miss a migrating thread,
	 * cpu_cgroup_attach() must not be possible after cgroup_exit() and it
	 * can't be removed from thread list, we hold ->siglock.
160 161 162
	 *
	 * If an exiting thread was already removed from thread list we rely on
	 * sched_autogroup_exit_task().
163
	 */
164
	for_each_thread(p, t)
165
		sched_move_task(t);
166

167 168 169 170 171 172 173
	unlock_task_sighand(p, &flags);
	autogroup_kref_put(prev);
}

/* Allocates GFP_KERNEL, cannot be called under any spinlock */
void sched_autogroup_create_attach(struct task_struct *p)
{
174
	struct autogroup *ag = autogroup_create();
175 176

	autogroup_move_group(p, ag);
L
Lucas De Marchi 已提交
177
	/* drop extra reference added by autogroup_create() */
178 179 180 181 182 183 184 185 186 187 188 189 190
	autogroup_kref_put(ag);
}
EXPORT_SYMBOL(sched_autogroup_create_attach);

/* Cannot be called under siglock.  Currently has no users */
void sched_autogroup_detach(struct task_struct *p)
{
	autogroup_move_group(p, &autogroup_default);
}
EXPORT_SYMBOL(sched_autogroup_detach);

void sched_autogroup_fork(struct signal_struct *sig)
{
191
	sig->autogroup = autogroup_task_get(current);
192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207
}

void sched_autogroup_exit(struct signal_struct *sig)
{
	autogroup_kref_put(sig->autogroup);
}

static int __init setup_autogroup(char *str)
{
	sysctl_sched_autogroup_enabled = 0;

	return 1;
}

__setup("noautogroup", setup_autogroup);

208 209 210 211 212 213
#ifdef CONFIG_PROC_FS

int proc_sched_autogroup_set_nice(struct task_struct *p, int nice)
{
	static unsigned long next = INITIAL_JIFFIES;
	struct autogroup *ag;
214
	unsigned long shares;
215 216
	int err;

217
	if (nice < MIN_NICE || nice > MAX_NICE)
218 219 220 221 222 223 224 225 226 227 228 229 230 231 232
		return -EINVAL;

	err = security_task_setnice(current, nice);
	if (err)
		return err;

	if (nice < 0 && !can_nice(current, nice))
		return -EPERM;

	/* this is a heavy operation taking global locks.. */
	if (!capable(CAP_SYS_ADMIN) && time_before(jiffies, next))
		return -EAGAIN;

	next = HZ / 10 + jiffies;
	ag = autogroup_task_get(p);
233
	shares = scale_load(sched_prio_to_weight[nice + 20]);
234 235

	down_write(&ag->lock);
236
	err = sched_group_set_shares(ag->tg, shares);
237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261
	if (!err)
		ag->nice = nice;
	up_write(&ag->lock);

	autogroup_kref_put(ag);

	return err;
}

void proc_sched_autogroup_show_task(struct task_struct *p, struct seq_file *m)
{
	struct autogroup *ag = autogroup_task_get(p);

	if (!task_group_is_autogroup(ag->tg))
		goto out;

	down_read(&ag->lock);
	seq_printf(m, "/autogroup-%ld nice %d\n", ag->id, ag->nice);
	up_read(&ag->lock);

out:
	autogroup_kref_put(ag);
}
#endif /* CONFIG_PROC_FS */

262
#ifdef CONFIG_SCHED_DEBUG
263
int autogroup_path(struct task_group *tg, char *buf, int buflen)
264
{
265
	if (!task_group_is_autogroup(tg))
266 267
		return 0;

268 269 270
	return snprintf(buf, buflen, "%s-%ld", "/autogroup", tg->autogroup->id);
}
#endif /* CONFIG_SCHED_DEBUG */