Subject: Provide individual CPU usage measurement based on idle time
From: Carsten Emde <C.Emde@osadl.org>
Date: Fri,  4 Sep 2015 20:41:40 +0100

The various methods to determine CPU usage and load have a number of
disadvantages (see also Documentation/cpu-load.txt), and a
straight-forward method to gain usage information about a particular CPU
is lacking. However, in the context of setting CPU affinity and
isolation, it is often required to monitor the effective usage ratio of
a CPU.

This patch adds an additional CPU usage measuring method that is based
on idle time processing. The data are available for every CPU in
/proc/idleruntime/cpuN/data in the format "<idletime> <runtime>". The
counters can be reset by writing to /proc/idleruntime/cpuN/reset.

To calculate the per-core CPU usage since the most recent reset, divide
the runtime by the sum of runtime plus idletime, e.g.

# for i in `ls -1d /proc/idleruntime/cpu* | sort -nk1.22`
> do
>   echo "$i: `awk '{ print (100.0*$2) / ($1+$2)"%" }' <$i/data`"
>   echo 1 >$i/reset
> done
/proc/idleruntime/cpu0: 72.0048%
/proc/idleruntime/cpu1: 5.49522%
/proc/idleruntime/cpu2: 0.27916%
/proc/idleruntime/cpu3: 32.3493%

In addition, summed up data of all present CPUs are available in
/proc/idleruntime/all in the same format as above. Thus, to calculate
the overall CPU usage since the most recent reset, the following command
may be used:
awk '{ print (100.0*$2) / ($1+$2)"%" }' </proc/idleruntime/all/data

To reset the counters althogether write to /proc/idleruntime/all/reset.

Signed-off-by: Carsten Emde <C.Emde@osadl.org>

---
 init/Kconfig                   |   28 +++++
 kernel/sched/Makefile          |    1 
 kernel/sched/core.c            |   31 +++++
 kernel/sched/cpu_idleruntime.c |  225 +++++++++++++++++++++++++++++++++++++++++
 kernel/sched/sched.h           |    8 +
 5 files changed, 293 insertions(+)

Index: linux-5.4.93-rt51/init/Kconfig
===================================================================
@ linux-5.4.93-rt51/init/Kconfig:464 @ config BSD_PROCESS_ACCT_V3
 	  for processing it. A preliminary version of these tools is available
 	  at <http://www.gnu.org/software/acct/>.
 
+config CPU_IDLERUNTIME
+	bool "Provide individual CPU usage measurement based on idle processing"
+	help
+	  If you say Y here, individual CPU usage data will be provided that are
+	  based on idle processing. The data are available for every CPU and for
+	  all present CPUs summed up in /proc/idleruntime/cpuN/data and
+	  /proc/idleruntime/all/data, respectively, in the format
+	  "<idletime> <runtime>". The counters can be reset by writing to
+	  /proc/idleruntime/cpuN/reset separately for every CPU and to
+	  /proc/idleruntime/all/reset for all present CPUs at once. To calculate
+	  the CPU usage since the most recent reset, the runtime must be devided
+	  by the sum of idletime plus runtime
+	    awk '{print (100.0*$2) / ($1+$2)"%"}' </proc/idleruntime/cpu0/data
+	  for every CPU or
+	    awk '{print (100.0*$2) / ($1+$2)"%"}' </proc/idleruntime/all/data
+	  for all CPUs altogether. The shell code snippet
+	  # for i in `ls -1d /proc/idleruntime/cpu* | sort -nk1.22`
+	  > do
+	  >   echo "$i: `awk '{ print (100.0*$2) / ($1+$2)"%" }' <$i/data`"
+	  >   echo 1 >$i/reset
+	  > done
+	  may produce
+	  /proc/idleruntime/cpu0: 72.0048%
+	  /proc/idleruntime/cpu1: 5.49522%
+	  /proc/idleruntime/cpu2: 0.27916%
+	  /proc/idleruntime/cpu3: 32.3493%
+	  on a four-core processor.
+
 config TASKSTATS
 	bool "Export task/process statistics through netlink"
 	depends on NET
Index: linux-5.4.93-rt51/kernel/sched/Makefile
===================================================================
--- linux-5.4.93-rt51.orig/kernel/sched/Makefile
+++ linux-5.4.93-rt51/kernel/sched/Makefile
@ linux-5.4.93-rt51/init/Kconfig:26 @ obj-y += wait.o wait_bit.o swait.o compl
 obj-$(CONFIG_SMP) += cpupri.o cpudeadline.o topology.o stop_task.o pelt.o
 obj-$(CONFIG_SCHED_AUTOGROUP) += autogroup.o
 obj-$(CONFIG_SCHEDSTATS) += stats.o
+obj-$(CONFIG_CPU_IDLERUNTIME) += cpu_idleruntime.o
 obj-$(CONFIG_SCHED_DEBUG) += debug.o
 obj-$(CONFIG_CGROUP_CPUACCT) += cpuacct.o
 obj-$(CONFIG_CPU_FREQ) += cpufreq.o
Index: linux-5.4.93-rt51/kernel/sched/core.c
===================================================================
--- linux-5.4.93-rt51.orig/kernel/sched/core.c
+++ linux-5.4.93-rt51/kernel/sched/core.c
@ linux-5.4.93-rt51/init/Kconfig:3385 @ prepare_task_switch(struct rq *rq, struc
 	fire_sched_out_preempt_notifiers(prev, next);
 	prepare_task(next);
 	prepare_arch_switch(next);
+#ifdef CONFIG_CPU_IDLERUNTIME
+	if (is_idle_task(next)) {
+		int cpu = raw_smp_processor_id();
+
+		if (per_cpu(idlestop, cpu)) {
+			unsigned long flags;
+
+			raw_spin_lock_irqsave(&per_cpu(idleruntime_lock, cpu),
+			    flags);
+			per_cpu(idlestart, cpu) = cpu_clock(cpu);
+			per_cpu(runtime, cpu) +=
+			    per_cpu(idlestart, cpu) - per_cpu(idlestop, cpu);
+			raw_spin_unlock_irqrestore(&per_cpu(idleruntime_lock,
+			    cpu), flags);
+		}
+	} else if (is_idle_task(prev)) {
+		int cpu = raw_smp_processor_id();
+
+		if (per_cpu(idlestart, cpu)) {
+			unsigned long flags;
+
+			raw_spin_lock_irqsave(&per_cpu(idleruntime_lock, cpu),
+			    flags);
+			per_cpu(idlestop, cpu) = cpu_clock(cpu);
+			per_cpu(idletime, cpu) +=
+			    per_cpu(idlestop, cpu) - per_cpu(idlestart, cpu);
+			raw_spin_unlock_irqrestore(&per_cpu(idleruntime_lock,
+			    cpu), flags);
+		}
+	}
+#endif
 }
 
 /**
Index: linux-5.4.93-rt51/kernel/sched/cpu_idleruntime.c
===================================================================
--- /dev/null
+++ linux-5.4.93-rt51/kernel/sched/cpu_idleruntime.c
@ linux-5.4.93-rt51/init/Kconfig:4 @
+/*
+   cpu_idleruntime.c: provide CPU usage data based on idle processing
+
+   Copyright (C) 2012,2015 Carsten Emde <C.Emde@osadl.org>
+
+   This program is free software; you can redistribute it and/or
+   modify it under the terms of the GNU General Public License
+   as published by the Free Software Foundation; either version 2
+   of the License, or (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with this program; if not, write to the Free Software
+   Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA.
+*/
+
+#include <linux/seq_file.h>
+#include <linux/proc_fs.h>
+#include <linux/cpu.h>
+
+#include "sched.h"
+
+DEFINE_PER_CPU(unsigned long long, idlestart);
+DEFINE_PER_CPU(unsigned long long, idlestop);
+DEFINE_PER_CPU(unsigned long long, idletime);
+DEFINE_PER_CPU(unsigned long long, runtime);
+DEFINE_PER_CPU(raw_spinlock_t, idleruntime_lock);
+
+static DEFINE_PER_CPU(struct proc_dir_entry *, idleruntime_dir);
+static struct proc_dir_entry *root_idleruntime_dir;
+
+static void idleruntime_get(unsigned long cpu, unsigned long long *cpuidletime,
+			    unsigned long long *cpuruntime)
+{
+	unsigned long long now;
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&per_cpu(idleruntime_lock, cpu), flags);
+
+	/* Update runtime counter */
+	now = cpu_clock(cpu);
+	if (is_idle_task(cpu_rq(cpu)->curr))
+		per_cpu(idletime, cpu) += now - per_cpu(idlestart, cpu);
+	else
+		per_cpu(runtime, cpu) += now - per_cpu(idlestop, cpu);
+
+	*cpuidletime = per_cpu(idletime, cpu);
+	*cpuruntime = per_cpu(runtime, cpu);
+
+	raw_spin_unlock_irqrestore(&per_cpu(idleruntime_lock, cpu), flags);
+
+}
+
+static void idleruntime_output(struct seq_file *m, unsigned long long idletime,
+		   unsigned long long runtime)
+{
+	seq_printf(m, "%llu %llu\n", idletime, runtime);
+}
+
+static int idleruntime_show(struct seq_file *m, void *v)
+{
+	unsigned long cpu = (unsigned long) m->private;
+	unsigned long long cpuidletime, cpuruntime;
+
+	idleruntime_get(cpu, &cpuidletime, &cpuruntime);
+	idleruntime_output(m, cpuidletime, cpuruntime);
+
+	return 0;
+}
+
+static int idleruntime_show_all(struct seq_file *m, void *v)
+{
+	unsigned long cpu;
+	unsigned long long total_idletime = 0ULL, total_runtime = 0ULL;
+
+	preempt_disable();
+
+	for_each_present_cpu(cpu) {
+		unsigned long long cpuidletime, cpuruntime;
+
+		idleruntime_get(cpu, &cpuidletime, &cpuruntime);
+		total_idletime += cpuidletime;
+		total_runtime += cpuruntime;
+	}
+
+	preempt_enable();
+
+	idleruntime_output(m, total_idletime, total_runtime);
+
+	return 0;
+}
+
+static inline void idleruntime_reset1(unsigned long cpu)
+{
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&per_cpu(idleruntime_lock, cpu), flags);
+	per_cpu(idletime, cpu) = per_cpu(runtime, cpu) = 0;
+	per_cpu(idlestart, cpu) =  per_cpu(idlestop, cpu) = cpu_clock(cpu);
+	raw_spin_unlock_irqrestore(&per_cpu(idleruntime_lock, cpu), flags);
+}
+
+static ssize_t idleruntime_reset(struct file *file, const char __user *buffer,
+				 size_t len, loff_t *offset)
+{
+	unsigned long cpu = (unsigned long) PDE_DATA(file_inode(file));
+
+	idleruntime_reset1(cpu);
+	return len;
+}
+
+static ssize_t idleruntime_reset_all(struct file *file,
+				    const char __user *buffer,
+				    size_t len, loff_t *offset)
+{
+	unsigned long cpu;
+
+	preempt_disable();
+
+	for_each_present_cpu(cpu)
+		idleruntime_reset1(cpu);
+
+	preempt_enable();
+
+	return len;
+}
+
+static int idleruntime_open_all(struct inode *inode, struct file *file)
+{
+	return single_open(file, idleruntime_show_all, PDE_DATA(inode));
+}
+
+static const struct file_operations idleruntime_all_fops = {
+	.open = idleruntime_open_all,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.write = idleruntime_reset_all,
+	.release = single_release,
+};
+
+static int idleruntime_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, idleruntime_show, PDE_DATA(inode));
+}
+
+static const struct file_operations idleruntime_fops = {
+	.open = idleruntime_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.write = idleruntime_reset,
+	.release = single_release,
+};
+
+static int setup_procfiles(unsigned int cpu)
+{
+	char name[32];
+	struct proc_dir_entry *idleruntime_cpudir = NULL;
+
+	if (root_idleruntime_dir) {
+		snprintf(name, sizeof(name), "cpu%d", cpu);
+		idleruntime_cpudir = proc_mkdir(name, root_idleruntime_dir);
+	}
+
+	if (idleruntime_cpudir) {
+		proc_create_data("data", S_IRUGO, idleruntime_cpudir,
+		    &idleruntime_fops, (void *) (long) cpu);
+		proc_create_data("reset", S_IWUGO, idleruntime_cpudir,
+		    &idleruntime_fops, (void *) (long) cpu);
+	}
+	per_cpu(idleruntime_dir, cpu) = idleruntime_cpudir;
+
+	return 0;
+}
+
+static int unset_procfiles(unsigned int cpu)
+{
+	struct proc_dir_entry *idleruntime_cpudir =
+	    per_cpu(idleruntime_dir, cpu);
+
+	if (idleruntime_cpudir) {
+		remove_proc_entry("reset", idleruntime_cpudir);
+		remove_proc_entry("data", idleruntime_cpudir);
+		proc_remove(idleruntime_cpudir);
+		per_cpu(idleruntime_dir, cpu) = NULL;
+	}
+
+	return 0;
+}
+
+static int __init idleruntime_init(void)
+{
+	root_idleruntime_dir = proc_mkdir("idleruntime", NULL);
+	if (root_idleruntime_dir) {
+		struct proc_dir_entry *idleruntime_alldir;
+		unsigned long cpu, cpus = 0;
+
+		for_each_possible_cpu(cpu) {
+			per_cpu(idlestart, cpu) =  per_cpu(idlestop, cpu) =
+			    cpu_clock(cpu);
+			raw_spin_lock_init(&per_cpu(idleruntime_lock, cpu));
+			cpus++;
+		}
+
+		if (cpus > 1) {
+			idleruntime_alldir = proc_mkdir("all",
+			    root_idleruntime_dir);
+			proc_create_data("data", S_IRUGO, idleruntime_alldir,
+			    &idleruntime_all_fops, NULL);
+			proc_create_data("reset", S_IWUGO, idleruntime_alldir,
+			    &idleruntime_all_fops, NULL);
+		}
+
+		cpuhp_setup_state(CPUHP_AP_ONLINE_DYN,
+				  "idleruntime:online",
+				  setup_procfiles,
+				  unset_procfiles);
+	}
+	return 0;
+}
+
+early_initcall(idleruntime_init);
Index: linux-5.4.93-rt51/kernel/sched/sched.h
===================================================================
--- linux-5.4.93-rt51.orig/kernel/sched/sched.h
+++ linux-5.4.93-rt51/kernel/sched/sched.h
@ linux-5.4.93-rt51/init/Kconfig:1269 @ this_rq_lock_irq(struct rq_flags *rf)
 	return rq;
 }
 
+#ifdef CONFIG_CPU_IDLERUNTIME
+extern DEFINE_PER_CPU(unsigned long long, idlestart);
+extern DEFINE_PER_CPU(unsigned long long, idlestop);
+extern DEFINE_PER_CPU(unsigned long long, idletime);
+extern DEFINE_PER_CPU(unsigned long long, runtime);
+extern DEFINE_PER_CPU(raw_spinlock_t, idleruntime_lock);
+#endif
+
 #ifdef CONFIG_NUMA
 enum numa_topology_type {
 	NUMA_DIRECT,