505 lines
11 KiB
C
505 lines
11 KiB
C
// SPDX-License-Identifier: GPL-2.0
|
|
/*
|
|
* CPU accounting code for task groups.
|
|
*
|
|
* Based on the work by Paul Menage (menage@google.com) and Balbir Singh
|
|
* (balbir@in.ibm.com).
|
|
*/
|
|
#include "sched.h"
|
|
#include <linux/sli.h>
|
|
|
|
/* Time spent by the tasks of the CPU accounting group executing in ... */
|
|
enum cpuacct_stat_index {
|
|
CPUACCT_STAT_USER, /* ... user mode */
|
|
CPUACCT_STAT_SYSTEM, /* ... kernel mode */
|
|
|
|
CPUACCT_STAT_NSTATS,
|
|
};
|
|
|
|
static const char * const cpuacct_stat_desc[] = {
|
|
[CPUACCT_STAT_USER] = "user",
|
|
[CPUACCT_STAT_SYSTEM] = "system",
|
|
};
|
|
|
|
struct cpuacct_usage {
|
|
u64 usages[CPUACCT_STAT_NSTATS];
|
|
};
|
|
|
|
/* track CPU usage of a group of tasks and its child groups */
|
|
struct cpuacct {
|
|
struct cgroup_subsys_state css;
|
|
/* cpuusage holds pointer to a u64-type object on every CPU */
|
|
struct cpuacct_usage __percpu *cpuusage;
|
|
struct kernel_cpustat __percpu *cpustat;
|
|
struct timespec64 uptime;
|
|
u64 idletime;
|
|
|
|
KABI_RESERVE(1);
|
|
KABI_RESERVE(2);
|
|
KABI_RESERVE(3);
|
|
KABI_RESERVE(4);
|
|
};
|
|
|
|
static inline struct cpuacct *css_ca(struct cgroup_subsys_state *css)
|
|
{
|
|
return css ? container_of(css, struct cpuacct, css) : NULL;
|
|
}
|
|
|
|
/* Return CPU accounting group to which this task belongs */
|
|
static inline struct cpuacct *task_ca(struct task_struct *tsk)
|
|
{
|
|
return css_ca(task_css(tsk, cpuacct_cgrp_id));
|
|
}
|
|
|
|
static inline struct cpuacct *parent_ca(struct cpuacct *ca)
|
|
{
|
|
return css_ca(ca->css.parent);
|
|
}
|
|
|
|
static DEFINE_PER_CPU(struct cpuacct_usage, root_cpuacct_cpuusage);
|
|
static struct cpuacct root_cpuacct = {
|
|
.cpustat = &kernel_cpustat,
|
|
.cpuusage = &root_cpuacct_cpuusage,
|
|
};
|
|
|
|
/* Create a new CPU accounting group */
|
|
static struct cgroup_subsys_state *
|
|
cpuacct_css_alloc(struct cgroup_subsys_state *parent_css)
|
|
{
|
|
struct cpuacct *ca;
|
|
u32 cpu;
|
|
|
|
if (!parent_css)
|
|
return &root_cpuacct.css;
|
|
|
|
ca = kzalloc(sizeof(*ca), GFP_KERNEL);
|
|
if (!ca)
|
|
goto out;
|
|
|
|
ca->cpuusage = alloc_percpu(struct cpuacct_usage);
|
|
if (!ca->cpuusage)
|
|
goto out_free_ca;
|
|
|
|
ca->cpustat = alloc_percpu(struct kernel_cpustat);
|
|
if (!ca->cpustat)
|
|
goto out_free_cpuusage;
|
|
|
|
ktime_get_boottime_ts64(&ca->uptime);
|
|
for_each_possible_cpu(cpu)
|
|
ca->idletime += (__force u64) kcpustat_cpu(cpu).cpustat[CPUTIME_IDLE];
|
|
|
|
return &ca->css;
|
|
|
|
out_free_cpuusage:
|
|
free_percpu(ca->cpuusage);
|
|
out_free_ca:
|
|
kfree(ca);
|
|
out:
|
|
return ERR_PTR(-ENOMEM);
|
|
}
|
|
|
|
/* Destroy an existing CPU accounting group */
|
|
static void cpuacct_css_free(struct cgroup_subsys_state *css)
|
|
{
|
|
struct cpuacct *ca = css_ca(css);
|
|
|
|
free_percpu(ca->cpustat);
|
|
free_percpu(ca->cpuusage);
|
|
kfree(ca);
|
|
}
|
|
|
|
static u64 cpuacct_cpuusage_read(struct cpuacct *ca, int cpu,
|
|
enum cpuacct_stat_index index)
|
|
{
|
|
struct cpuacct_usage *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
|
|
u64 data;
|
|
|
|
/*
|
|
* We allow index == CPUACCT_STAT_NSTATS here to read
|
|
* the sum of suages.
|
|
*/
|
|
BUG_ON(index > CPUACCT_STAT_NSTATS);
|
|
|
|
#ifndef CONFIG_64BIT
|
|
/*
|
|
* Take rq->lock to make 64-bit read safe on 32-bit platforms.
|
|
*/
|
|
raw_spin_rq_lock_irq(cpu_rq(cpu));
|
|
#endif
|
|
|
|
if (index == CPUACCT_STAT_NSTATS) {
|
|
int i = 0;
|
|
|
|
data = 0;
|
|
for (i = 0; i < CPUACCT_STAT_NSTATS; i++)
|
|
data += cpuusage->usages[i];
|
|
} else {
|
|
data = cpuusage->usages[index];
|
|
}
|
|
|
|
#ifndef CONFIG_64BIT
|
|
raw_spin_rq_unlock_irq(cpu_rq(cpu));
|
|
#endif
|
|
|
|
return data;
|
|
}
|
|
|
|
static void cpuacct_cpuusage_write(struct cpuacct *ca, int cpu, u64 val)
|
|
{
|
|
struct cpuacct_usage *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
|
|
int i;
|
|
|
|
#ifndef CONFIG_64BIT
|
|
/*
|
|
* Take rq->lock to make 64-bit write safe on 32-bit platforms.
|
|
*/
|
|
raw_spin_rq_lock_irq(cpu_rq(cpu));
|
|
#endif
|
|
|
|
for (i = 0; i < CPUACCT_STAT_NSTATS; i++)
|
|
cpuusage->usages[i] = val;
|
|
|
|
#ifndef CONFIG_64BIT
|
|
raw_spin_rq_unlock_irq(cpu_rq(cpu));
|
|
#endif
|
|
}
|
|
|
|
/* Return total CPU usage (in nanoseconds) of a group */
|
|
static u64 __cpuusage_read(struct cgroup_subsys_state *css,
|
|
enum cpuacct_stat_index index)
|
|
{
|
|
struct cpuacct *ca = css_ca(css);
|
|
u64 totalcpuusage = 0;
|
|
int i;
|
|
|
|
for_each_possible_cpu(i)
|
|
totalcpuusage += cpuacct_cpuusage_read(ca, i, index);
|
|
|
|
return totalcpuusage;
|
|
}
|
|
|
|
static u64 cpuusage_user_read(struct cgroup_subsys_state *css,
|
|
struct cftype *cft)
|
|
{
|
|
return __cpuusage_read(css, CPUACCT_STAT_USER);
|
|
}
|
|
|
|
static u64 cpuusage_sys_read(struct cgroup_subsys_state *css,
|
|
struct cftype *cft)
|
|
{
|
|
return __cpuusage_read(css, CPUACCT_STAT_SYSTEM);
|
|
}
|
|
|
|
static u64 cpuusage_read(struct cgroup_subsys_state *css, struct cftype *cft)
|
|
{
|
|
return __cpuusage_read(css, CPUACCT_STAT_NSTATS);
|
|
}
|
|
|
|
static int cpuusage_write(struct cgroup_subsys_state *css, struct cftype *cft,
|
|
u64 val)
|
|
{
|
|
struct cpuacct *ca = css_ca(css);
|
|
int cpu;
|
|
|
|
/*
|
|
* Only allow '0' here to do a reset.
|
|
*/
|
|
if (val)
|
|
return -EINVAL;
|
|
|
|
for_each_possible_cpu(cpu)
|
|
cpuacct_cpuusage_write(ca, cpu, 0);
|
|
|
|
return 0;
|
|
}
|
|
|
|
static int __cpuacct_percpu_seq_show(struct seq_file *m,
|
|
enum cpuacct_stat_index index)
|
|
{
|
|
struct cpuacct *ca = css_ca(seq_css(m));
|
|
u64 percpu;
|
|
int i;
|
|
|
|
for_each_possible_cpu(i) {
|
|
percpu = cpuacct_cpuusage_read(ca, i, index);
|
|
seq_printf(m, "%llu ", (unsigned long long) percpu);
|
|
}
|
|
seq_printf(m, "\n");
|
|
return 0;
|
|
}
|
|
|
|
static int cpuacct_percpu_user_seq_show(struct seq_file *m, void *V)
|
|
{
|
|
return __cpuacct_percpu_seq_show(m, CPUACCT_STAT_USER);
|
|
}
|
|
|
|
static int cpuacct_percpu_sys_seq_show(struct seq_file *m, void *V)
|
|
{
|
|
return __cpuacct_percpu_seq_show(m, CPUACCT_STAT_SYSTEM);
|
|
}
|
|
|
|
static int cpuacct_percpu_seq_show(struct seq_file *m, void *V)
|
|
{
|
|
return __cpuacct_percpu_seq_show(m, CPUACCT_STAT_NSTATS);
|
|
}
|
|
|
|
static int cpuacct_all_seq_show(struct seq_file *m, void *V)
|
|
{
|
|
struct cpuacct *ca = css_ca(seq_css(m));
|
|
int index;
|
|
int cpu;
|
|
|
|
seq_puts(m, "cpu");
|
|
for (index = 0; index < CPUACCT_STAT_NSTATS; index++)
|
|
seq_printf(m, " %s", cpuacct_stat_desc[index]);
|
|
seq_puts(m, "\n");
|
|
|
|
for_each_possible_cpu(cpu) {
|
|
struct cpuacct_usage *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
|
|
|
|
seq_printf(m, "%d", cpu);
|
|
|
|
for (index = 0; index < CPUACCT_STAT_NSTATS; index++) {
|
|
#ifndef CONFIG_64BIT
|
|
/*
|
|
* Take rq->lock to make 64-bit read safe on 32-bit
|
|
* platforms.
|
|
*/
|
|
raw_spin_rq_lock_irq(cpu_rq(cpu));
|
|
#endif
|
|
|
|
seq_printf(m, " %llu", cpuusage->usages[index]);
|
|
|
|
#ifndef CONFIG_64BIT
|
|
raw_spin_rq_unlock_irq(cpu_rq(cpu));
|
|
#endif
|
|
}
|
|
seq_puts(m, "\n");
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
static int cpuacct_stats_show(struct seq_file *sf, void *v)
|
|
{
|
|
struct cpuacct *ca = css_ca(seq_css(sf));
|
|
s64 val[CPUACCT_STAT_NSTATS];
|
|
int cpu;
|
|
int stat;
|
|
|
|
memset(val, 0, sizeof(val));
|
|
for_each_possible_cpu(cpu) {
|
|
u64 *cpustat = per_cpu_ptr(ca->cpustat, cpu)->cpustat;
|
|
|
|
val[CPUACCT_STAT_USER] += cpustat[CPUTIME_USER];
|
|
val[CPUACCT_STAT_USER] += cpustat[CPUTIME_NICE];
|
|
val[CPUACCT_STAT_SYSTEM] += cpustat[CPUTIME_SYSTEM];
|
|
val[CPUACCT_STAT_SYSTEM] += cpustat[CPUTIME_IRQ];
|
|
val[CPUACCT_STAT_SYSTEM] += cpustat[CPUTIME_SOFTIRQ];
|
|
}
|
|
|
|
for (stat = 0; stat < CPUACCT_STAT_NSTATS; stat++) {
|
|
seq_printf(sf, "%s %lld\n",
|
|
cpuacct_stat_desc[stat],
|
|
(long long)nsec_to_clock_t(val[stat]));
|
|
}
|
|
|
|
return 0;
|
|
}
|
|
|
|
static int cpuacct_uptime_show_comm(struct seq_file *sf, void *v, struct cpuacct *ca)
|
|
{
|
|
struct timespec64 uptime, idle, curr;
|
|
u64 idletime = 0;
|
|
u32 cpu, rem;
|
|
|
|
ktime_get_boottime_ts64(&curr);
|
|
uptime = timespec64_sub(curr, ca->uptime);
|
|
|
|
for_each_possible_cpu(cpu)
|
|
idletime += (__force u64) kcpustat_cpu(cpu).cpustat[CPUTIME_IDLE];
|
|
|
|
idle.tv_sec = div_u64_rem(idletime - ca->idletime, NSEC_PER_SEC, &rem);
|
|
idle.tv_nsec = rem;
|
|
seq_printf(sf, "%lu.%02lu %lu.%02lu\n",
|
|
(unsigned long) uptime.tv_sec,
|
|
(uptime.tv_nsec / (NSEC_PER_SEC / 100)),
|
|
(unsigned long) idle.tv_sec,
|
|
(idle.tv_nsec / (NSEC_PER_SEC / 100)));
|
|
return 0;
|
|
}
|
|
|
|
#ifdef CONFIG_CGROUP_SLI
|
|
static int cpuacct_sli_show(struct seq_file *sf, void *v)
|
|
{
|
|
struct cgroup *cgrp = seq_css(sf)->cgroup;
|
|
|
|
return sli_schedlat_stat_show(sf, cgrp);
|
|
}
|
|
|
|
static int cpuacct_sli_max_show(struct seq_file *sf, void *v)
|
|
{
|
|
struct cgroup *cgrp = seq_css(sf)->cgroup;
|
|
|
|
return sli_schedlat_max_show(sf, cgrp);
|
|
}
|
|
#endif
|
|
|
|
int cpuacct_cgroupfs_uptime_show(struct seq_file *m, void *v)
|
|
{
|
|
int ret;
|
|
struct cgroup_subsys_state *css;
|
|
struct cpuacct *ca;
|
|
|
|
css = cgroupfs_get_parent_role_cgroup(current,
|
|
CGROUPFS_CGROUP_ROLE_POD_GROUPS, cpuacct_cgrp_id);
|
|
ca = css_ca(css);
|
|
ret = cpuacct_uptime_show_comm(m, v, ca);
|
|
css_put(css);
|
|
|
|
return ret;
|
|
}
|
|
|
|
int cpuacct_cgroupfs_cpu_usage(struct cgroup_subsys_state *css, int cpu, u64 *sys, u64 *user)
|
|
{
|
|
struct cpuacct *ca = css_ca(css);
|
|
*user = cpuacct_cpuusage_read(ca, cpu, CPUACCT_STAT_USER);
|
|
*sys = cpuacct_cpuusage_read(ca, cpu, CPUACCT_STAT_SYSTEM);
|
|
return 0;
|
|
}
|
|
|
|
static int cpuacct_uptime_show(struct seq_file *sf, void *v)
|
|
{
|
|
struct cpuacct *ca = css_ca(seq_css(sf));
|
|
return cpuacct_uptime_show_comm(sf, v, ca);
|
|
}
|
|
|
|
static struct cftype files[] = {
|
|
{
|
|
.name = "usage",
|
|
.read_u64 = cpuusage_read,
|
|
.write_u64 = cpuusage_write,
|
|
},
|
|
{
|
|
.name = "usage_user",
|
|
.read_u64 = cpuusage_user_read,
|
|
},
|
|
{
|
|
.name = "usage_sys",
|
|
.read_u64 = cpuusage_sys_read,
|
|
},
|
|
{
|
|
.name = "usage_percpu",
|
|
.seq_show = cpuacct_percpu_seq_show,
|
|
},
|
|
{
|
|
.name = "usage_percpu_user",
|
|
.seq_show = cpuacct_percpu_user_seq_show,
|
|
},
|
|
{
|
|
.name = "usage_percpu_sys",
|
|
.seq_show = cpuacct_percpu_sys_seq_show,
|
|
},
|
|
{
|
|
.name = "usage_all",
|
|
.seq_show = cpuacct_all_seq_show,
|
|
},
|
|
{
|
|
.name = "stat",
|
|
.seq_show = cpuacct_stats_show,
|
|
},
|
|
{
|
|
.name = "uptime",
|
|
.seq_show = cpuacct_uptime_show,
|
|
},
|
|
{
|
|
.name = "mbuf",
|
|
.flags = CFTYPE_NOT_ON_ROOT,
|
|
.seq_show = cgroup_mbuf_show,
|
|
.seq_start = cgroup_mbuf_start,
|
|
.seq_next = cgroup_mbuf_next,
|
|
.seq_stop = cgroup_mbuf_stop,
|
|
},
|
|
#ifdef CONFIG_PSI
|
|
{
|
|
.name = "cpu.pressure",
|
|
.flags = CFTYPE_NO_PREFIX,
|
|
.seq_show = cgroup_cpu_pressure_show,
|
|
.write = cgroup_cpu_pressure_write,
|
|
.poll = cgroup_pressure_poll,
|
|
.release = cgroup_pressure_release,
|
|
},
|
|
#endif /* CONFIG_PSI */
|
|
#ifdef CONFIG_CGROUP_SLI
|
|
{
|
|
.name = "sli",
|
|
.flags = CFTYPE_NOT_ON_ROOT,
|
|
.seq_show = cpuacct_sli_show,
|
|
},
|
|
{
|
|
.name = "sli_max",
|
|
.flags = CFTYPE_NOT_ON_ROOT,
|
|
.seq_show = cpuacct_sli_max_show,
|
|
},
|
|
{
|
|
.name = "sli.control",
|
|
.write = cgroup_sli_control_write,
|
|
.seq_show = cgroup_sli_control_show,
|
|
},
|
|
{
|
|
.name = "sli.monitor",
|
|
.flags = CFTYPE_NOT_ON_ROOT,
|
|
.open = cgroup_sli_monitor_open,
|
|
.seq_show = cgroup_sli_monitor_show,
|
|
.seq_start = cgroup_sli_monitor_start,
|
|
.seq_next = cgroup_sli_monitor_next,
|
|
.seq_stop = cgroup_sli_monitor_stop,
|
|
.poll = cgroup_sli_monitor_poll,
|
|
},
|
|
#endif
|
|
{ } /* terminate */
|
|
};
|
|
|
|
/*
|
|
* charge this task's execution time to its accounting group.
|
|
*
|
|
* called with rq->lock held.
|
|
*/
|
|
void cpuacct_charge(struct task_struct *tsk, u64 cputime)
|
|
{
|
|
struct cpuacct *ca;
|
|
int index = CPUACCT_STAT_SYSTEM;
|
|
struct pt_regs *regs = task_pt_regs(tsk);
|
|
|
|
if (regs && user_mode(regs))
|
|
index = CPUACCT_STAT_USER;
|
|
|
|
rcu_read_lock();
|
|
|
|
for (ca = task_ca(tsk); ca; ca = parent_ca(ca))
|
|
this_cpu_ptr(ca->cpuusage)->usages[index] += cputime;
|
|
|
|
rcu_read_unlock();
|
|
}
|
|
|
|
/*
|
|
* Add user/system time to cpuacct.
|
|
*
|
|
* Note: it's the caller that updates the account of the root cgroup.
|
|
*/
|
|
void cpuacct_account_field(struct task_struct *tsk, int index, u64 val)
|
|
{
|
|
struct cpuacct *ca;
|
|
|
|
rcu_read_lock();
|
|
for (ca = task_ca(tsk); ca != &root_cpuacct; ca = parent_ca(ca))
|
|
this_cpu_ptr(ca->cpustat)->cpustat[index] += val;
|
|
rcu_read_unlock();
|
|
}
|
|
|
|
struct cgroup_subsys cpuacct_cgrp_subsys = {
|
|
.css_alloc = cpuacct_css_alloc,
|
|
.css_free = cpuacct_css_free,
|
|
.legacy_cftypes = files,
|
|
.early_init = true,
|
|
};
|