samples/bpf: Add program for CPU state statistics
CPU is active when have running tasks on it and CPUFreq governor can
select different operating points (OPP) according to different workload;
we use 'pstate' to present CPU state which have running tasks with one
specific OPP. On the other hand, CPU is idle which only idle task on
it, CPUIdle governor can select one specific idle state to power off
hardware logics; we use 'cstate' to present CPU idle state.
Based on trace events 'cpu_idle' and 'cpu_frequency' we can accomplish
the duration statistics for every state. Every time when CPU enters
into or exits from idle states, the trace event 'cpu_idle' is recorded;
trace event 'cpu_frequency' records the event for CPU OPP changing, so
it's easily to know how long time the CPU stays in the specified OPP,
and the CPU must be not in any idle state.
This patch is to utilize the mentioned trace events for pstate and
cstate statistics. To achieve more accurate profiling data, the program
uses below sequence to insure CPU running/idle time aren't missed:
- Before profiling the user space program wakes up all CPUs for once, so
can avoid to missing account time for CPU staying in idle state for
long time; the program forces to set 'scaling_max_freq' to lowest
frequency and then restore 'scaling_max_freq' to highest frequency,
this can ensure the frequency to be set to lowest frequency and later
after start to run workload the frequency can be easily to be changed
to higher frequency;
- User space program reads map data and update statistics for every 5s,
so this is same with other sample bpf programs for avoiding big
overload introduced by bpf program self;
- When send signal to terminate program, the signal handler wakes up
all CPUs, set lowest frequency and restore highest frequency to
'scaling_max_freq'; this is exactly same with the first step so
avoid to missing account CPU pstate and cstate time during last
stage. Finally it reports the latest statistics.
The program has been tested on Hikey board with octa CA53 CPUs, below
is one example for statistics result, the format mainly follows up
Jesper Dangaard Brouer suggestion.
Jesper reminds to 'get printf to pretty print with thousands separators
use %' and setlocale(LC_NUMERIC, "en_US")', tried three different arm64
GCC toolchains (5.4.0 20160609, 6.2.1 20161016, 6.3.0 20170516) but all
of them cannot support printf flag character %' on arm64 platform, so go
back print number without grouping mode.
CPU states statistics:
state(ms) cstate-0 cstate-1 cstate-2 pstate-0 pstate-1 pstate-2 pstate-3 pstate-4
CPU-0 767 6111 111863 561 31 756 853 190
CPU-1 241 10606 107956 484 125 646 990 85
CPU-2 413 19721 98735 636 84 696 757 89
CPU-3 84 11711 79989 17516 909 4811 5773 341
CPU-4 152 19610 98229 444 53 649 708 1283
CPU-5 185 8781 108697 666 91 671 677 1365
CPU-6 157 21964 95825 581 67 566 684 1284
CPU-7 125 15238 102704 398 20 665 786 1197
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-02-26 04:19:12 +03:00
|
|
|
// SPDX-License-Identifier: GPL-2.0
|
|
|
|
|
|
|
|
#define _GNU_SOURCE
|
|
|
|
#include <errno.h>
|
|
|
|
#include <stdio.h>
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <signal.h>
|
|
|
|
#include <sched.h>
|
|
|
|
#include <string.h>
|
|
|
|
#include <unistd.h>
|
|
|
|
#include <fcntl.h>
|
|
|
|
#include <locale.h>
|
|
|
|
#include <sys/types.h>
|
|
|
|
#include <sys/stat.h>
|
|
|
|
#include <sys/time.h>
|
|
|
|
#include <sys/resource.h>
|
|
|
|
#include <sys/wait.h>
|
|
|
|
|
2018-05-15 08:35:02 +03:00
|
|
|
#include <bpf/bpf.h>
|
2020-08-23 11:53:34 +03:00
|
|
|
#include <bpf/libbpf.h>
|
|
|
|
|
|
|
|
static int cstate_map_fd, pstate_map_fd;
|
samples/bpf: Add program for CPU state statistics
CPU is active when have running tasks on it and CPUFreq governor can
select different operating points (OPP) according to different workload;
we use 'pstate' to present CPU state which have running tasks with one
specific OPP. On the other hand, CPU is idle which only idle task on
it, CPUIdle governor can select one specific idle state to power off
hardware logics; we use 'cstate' to present CPU idle state.
Based on trace events 'cpu_idle' and 'cpu_frequency' we can accomplish
the duration statistics for every state. Every time when CPU enters
into or exits from idle states, the trace event 'cpu_idle' is recorded;
trace event 'cpu_frequency' records the event for CPU OPP changing, so
it's easily to know how long time the CPU stays in the specified OPP,
and the CPU must be not in any idle state.
This patch is to utilize the mentioned trace events for pstate and
cstate statistics. To achieve more accurate profiling data, the program
uses below sequence to insure CPU running/idle time aren't missed:
- Before profiling the user space program wakes up all CPUs for once, so
can avoid to missing account time for CPU staying in idle state for
long time; the program forces to set 'scaling_max_freq' to lowest
frequency and then restore 'scaling_max_freq' to highest frequency,
this can ensure the frequency to be set to lowest frequency and later
after start to run workload the frequency can be easily to be changed
to higher frequency;
- User space program reads map data and update statistics for every 5s,
so this is same with other sample bpf programs for avoiding big
overload introduced by bpf program self;
- When send signal to terminate program, the signal handler wakes up
all CPUs, set lowest frequency and restore highest frequency to
'scaling_max_freq'; this is exactly same with the first step so
avoid to missing account CPU pstate and cstate time during last
stage. Finally it reports the latest statistics.
The program has been tested on Hikey board with octa CA53 CPUs, below
is one example for statistics result, the format mainly follows up
Jesper Dangaard Brouer suggestion.
Jesper reminds to 'get printf to pretty print with thousands separators
use %' and setlocale(LC_NUMERIC, "en_US")', tried three different arm64
GCC toolchains (5.4.0 20160609, 6.2.1 20161016, 6.3.0 20170516) but all
of them cannot support printf flag character %' on arm64 platform, so go
back print number without grouping mode.
CPU states statistics:
state(ms) cstate-0 cstate-1 cstate-2 pstate-0 pstate-1 pstate-2 pstate-3 pstate-4
CPU-0 767 6111 111863 561 31 756 853 190
CPU-1 241 10606 107956 484 125 646 990 85
CPU-2 413 19721 98735 636 84 696 757 89
CPU-3 84 11711 79989 17516 909 4811 5773 341
CPU-4 152 19610 98229 444 53 649 708 1283
CPU-5 185 8781 108697 666 91 671 677 1365
CPU-6 157 21964 95825 581 67 566 684 1284
CPU-7 125 15238 102704 398 20 665 786 1197
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-02-26 04:19:12 +03:00
|
|
|
|
|
|
|
#define MAX_CPU 8
|
|
|
|
#define MAX_PSTATE_ENTRIES 5
|
|
|
|
#define MAX_CSTATE_ENTRIES 3
|
|
|
|
#define MAX_STARS 40
|
|
|
|
|
|
|
|
#define CPUFREQ_MAX_SYSFS_PATH "/sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq"
|
|
|
|
#define CPUFREQ_LOWEST_FREQ "208000"
|
|
|
|
#define CPUFREQ_HIGHEST_FREQ "12000000"
|
|
|
|
|
|
|
|
struct cpu_stat_data {
|
|
|
|
unsigned long cstate[MAX_CSTATE_ENTRIES];
|
|
|
|
unsigned long pstate[MAX_PSTATE_ENTRIES];
|
|
|
|
};
|
|
|
|
|
|
|
|
static struct cpu_stat_data stat_data[MAX_CPU];
|
|
|
|
|
|
|
|
static void cpu_stat_print(void)
|
|
|
|
{
|
|
|
|
int i, j;
|
|
|
|
char state_str[sizeof("cstate-9")];
|
|
|
|
struct cpu_stat_data *data;
|
|
|
|
|
|
|
|
/* Clear screen */
|
|
|
|
printf("\033[2J");
|
|
|
|
|
|
|
|
/* Header */
|
|
|
|
printf("\nCPU states statistics:\n");
|
|
|
|
printf("%-10s ", "state(ms)");
|
|
|
|
|
|
|
|
for (i = 0; i < MAX_CSTATE_ENTRIES; i++) {
|
|
|
|
sprintf(state_str, "cstate-%d", i);
|
|
|
|
printf("%-11s ", state_str);
|
|
|
|
}
|
|
|
|
|
|
|
|
for (i = 0; i < MAX_PSTATE_ENTRIES; i++) {
|
|
|
|
sprintf(state_str, "pstate-%d", i);
|
|
|
|
printf("%-11s ", state_str);
|
|
|
|
}
|
|
|
|
|
|
|
|
printf("\n");
|
|
|
|
|
|
|
|
for (j = 0; j < MAX_CPU; j++) {
|
|
|
|
data = &stat_data[j];
|
|
|
|
|
|
|
|
printf("CPU-%-6d ", j);
|
|
|
|
for (i = 0; i < MAX_CSTATE_ENTRIES; i++)
|
|
|
|
printf("%-11ld ", data->cstate[i] / 1000000);
|
|
|
|
|
|
|
|
for (i = 0; i < MAX_PSTATE_ENTRIES; i++)
|
|
|
|
printf("%-11ld ", data->pstate[i] / 1000000);
|
|
|
|
|
|
|
|
printf("\n");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void cpu_stat_update(int cstate_fd, int pstate_fd)
|
|
|
|
{
|
|
|
|
unsigned long key, value;
|
|
|
|
int c, i;
|
|
|
|
|
|
|
|
for (c = 0; c < MAX_CPU; c++) {
|
|
|
|
for (i = 0; i < MAX_CSTATE_ENTRIES; i++) {
|
|
|
|
key = c * MAX_CSTATE_ENTRIES + i;
|
|
|
|
bpf_map_lookup_elem(cstate_fd, &key, &value);
|
|
|
|
stat_data[c].cstate[i] = value;
|
|
|
|
}
|
|
|
|
|
|
|
|
for (i = 0; i < MAX_PSTATE_ENTRIES; i++) {
|
|
|
|
key = c * MAX_PSTATE_ENTRIES + i;
|
|
|
|
bpf_map_lookup_elem(pstate_fd, &key, &value);
|
|
|
|
stat_data[c].pstate[i] = value;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This function is copied from 'idlestat' tool function
|
|
|
|
* idlestat_wake_all() in idlestate.c.
|
|
|
|
*
|
|
|
|
* It sets the self running task affinity to cpus one by one so can wake up
|
|
|
|
* the specific CPU to handle scheduling; this results in all cpus can be
|
|
|
|
* waken up once and produce ftrace event 'trace_cpu_idle'.
|
|
|
|
*/
|
|
|
|
static int cpu_stat_inject_cpu_idle_event(void)
|
|
|
|
{
|
|
|
|
int rcpu, i, ret;
|
|
|
|
cpu_set_t cpumask;
|
|
|
|
cpu_set_t original_cpumask;
|
|
|
|
|
|
|
|
ret = sysconf(_SC_NPROCESSORS_CONF);
|
|
|
|
if (ret < 0)
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
rcpu = sched_getcpu();
|
|
|
|
if (rcpu < 0)
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
/* Keep track of the CPUs we will run on */
|
|
|
|
sched_getaffinity(0, sizeof(original_cpumask), &original_cpumask);
|
|
|
|
|
|
|
|
for (i = 0; i < ret; i++) {
|
|
|
|
|
|
|
|
/* Pointless to wake up ourself */
|
|
|
|
if (i == rcpu)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
/* Pointless to wake CPUs we will not run on */
|
|
|
|
if (!CPU_ISSET(i, &original_cpumask))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
CPU_ZERO(&cpumask);
|
|
|
|
CPU_SET(i, &cpumask);
|
|
|
|
|
|
|
|
sched_setaffinity(0, sizeof(cpumask), &cpumask);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Enable all the CPUs of the original mask */
|
|
|
|
sched_setaffinity(0, sizeof(original_cpumask), &original_cpumask);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* It's possible to have no any frequency change for long time and cannot
|
|
|
|
* get ftrace event 'trace_cpu_frequency' for long period, this introduces
|
|
|
|
* big deviation for pstate statistics.
|
|
|
|
*
|
|
|
|
* To solve this issue, below code forces to set 'scaling_max_freq' to 208MHz
|
|
|
|
* for triggering ftrace event 'trace_cpu_frequency' and then recovery back to
|
|
|
|
* the maximum frequency value 1.2GHz.
|
|
|
|
*/
|
|
|
|
static int cpu_stat_inject_cpu_frequency_event(void)
|
|
|
|
{
|
|
|
|
int len, fd;
|
|
|
|
|
|
|
|
fd = open(CPUFREQ_MAX_SYSFS_PATH, O_WRONLY);
|
|
|
|
if (fd < 0) {
|
|
|
|
printf("failed to open scaling_max_freq, errno=%d\n", errno);
|
|
|
|
return fd;
|
|
|
|
}
|
|
|
|
|
|
|
|
len = write(fd, CPUFREQ_LOWEST_FREQ, strlen(CPUFREQ_LOWEST_FREQ));
|
|
|
|
if (len < 0) {
|
|
|
|
printf("failed to open scaling_max_freq, errno=%d\n", errno);
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
len = write(fd, CPUFREQ_HIGHEST_FREQ, strlen(CPUFREQ_HIGHEST_FREQ));
|
|
|
|
if (len < 0) {
|
|
|
|
printf("failed to open scaling_max_freq, errno=%d\n", errno);
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
err:
|
|
|
|
close(fd);
|
|
|
|
return len;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void int_exit(int sig)
|
|
|
|
{
|
|
|
|
cpu_stat_inject_cpu_idle_event();
|
|
|
|
cpu_stat_inject_cpu_frequency_event();
|
2020-08-23 11:53:34 +03:00
|
|
|
cpu_stat_update(cstate_map_fd, pstate_map_fd);
|
samples/bpf: Add program for CPU state statistics
CPU is active when have running tasks on it and CPUFreq governor can
select different operating points (OPP) according to different workload;
we use 'pstate' to present CPU state which have running tasks with one
specific OPP. On the other hand, CPU is idle which only idle task on
it, CPUIdle governor can select one specific idle state to power off
hardware logics; we use 'cstate' to present CPU idle state.
Based on trace events 'cpu_idle' and 'cpu_frequency' we can accomplish
the duration statistics for every state. Every time when CPU enters
into or exits from idle states, the trace event 'cpu_idle' is recorded;
trace event 'cpu_frequency' records the event for CPU OPP changing, so
it's easily to know how long time the CPU stays in the specified OPP,
and the CPU must be not in any idle state.
This patch is to utilize the mentioned trace events for pstate and
cstate statistics. To achieve more accurate profiling data, the program
uses below sequence to insure CPU running/idle time aren't missed:
- Before profiling the user space program wakes up all CPUs for once, so
can avoid to missing account time for CPU staying in idle state for
long time; the program forces to set 'scaling_max_freq' to lowest
frequency and then restore 'scaling_max_freq' to highest frequency,
this can ensure the frequency to be set to lowest frequency and later
after start to run workload the frequency can be easily to be changed
to higher frequency;
- User space program reads map data and update statistics for every 5s,
so this is same with other sample bpf programs for avoiding big
overload introduced by bpf program self;
- When send signal to terminate program, the signal handler wakes up
all CPUs, set lowest frequency and restore highest frequency to
'scaling_max_freq'; this is exactly same with the first step so
avoid to missing account CPU pstate and cstate time during last
stage. Finally it reports the latest statistics.
The program has been tested on Hikey board with octa CA53 CPUs, below
is one example for statistics result, the format mainly follows up
Jesper Dangaard Brouer suggestion.
Jesper reminds to 'get printf to pretty print with thousands separators
use %' and setlocale(LC_NUMERIC, "en_US")', tried three different arm64
GCC toolchains (5.4.0 20160609, 6.2.1 20161016, 6.3.0 20170516) but all
of them cannot support printf flag character %' on arm64 platform, so go
back print number without grouping mode.
CPU states statistics:
state(ms) cstate-0 cstate-1 cstate-2 pstate-0 pstate-1 pstate-2 pstate-3 pstate-4
CPU-0 767 6111 111863 561 31 756 853 190
CPU-1 241 10606 107956 484 125 646 990 85
CPU-2 413 19721 98735 636 84 696 757 89
CPU-3 84 11711 79989 17516 909 4811 5773 341
CPU-4 152 19610 98229 444 53 649 708 1283
CPU-5 185 8781 108697 666 91 671 677 1365
CPU-6 157 21964 95825 581 67 566 684 1284
CPU-7 125 15238 102704 398 20 665 786 1197
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-02-26 04:19:12 +03:00
|
|
|
cpu_stat_print();
|
|
|
|
exit(0);
|
|
|
|
}
|
|
|
|
|
|
|
|
int main(int argc, char **argv)
|
|
|
|
{
|
2020-08-23 11:53:34 +03:00
|
|
|
struct bpf_link *link = NULL;
|
|
|
|
struct bpf_program *prog;
|
|
|
|
struct bpf_object *obj;
|
samples/bpf: Add program for CPU state statistics
CPU is active when have running tasks on it and CPUFreq governor can
select different operating points (OPP) according to different workload;
we use 'pstate' to present CPU state which have running tasks with one
specific OPP. On the other hand, CPU is idle which only idle task on
it, CPUIdle governor can select one specific idle state to power off
hardware logics; we use 'cstate' to present CPU idle state.
Based on trace events 'cpu_idle' and 'cpu_frequency' we can accomplish
the duration statistics for every state. Every time when CPU enters
into or exits from idle states, the trace event 'cpu_idle' is recorded;
trace event 'cpu_frequency' records the event for CPU OPP changing, so
it's easily to know how long time the CPU stays in the specified OPP,
and the CPU must be not in any idle state.
This patch is to utilize the mentioned trace events for pstate and
cstate statistics. To achieve more accurate profiling data, the program
uses below sequence to insure CPU running/idle time aren't missed:
- Before profiling the user space program wakes up all CPUs for once, so
can avoid to missing account time for CPU staying in idle state for
long time; the program forces to set 'scaling_max_freq' to lowest
frequency and then restore 'scaling_max_freq' to highest frequency,
this can ensure the frequency to be set to lowest frequency and later
after start to run workload the frequency can be easily to be changed
to higher frequency;
- User space program reads map data and update statistics for every 5s,
so this is same with other sample bpf programs for avoiding big
overload introduced by bpf program self;
- When send signal to terminate program, the signal handler wakes up
all CPUs, set lowest frequency and restore highest frequency to
'scaling_max_freq'; this is exactly same with the first step so
avoid to missing account CPU pstate and cstate time during last
stage. Finally it reports the latest statistics.
The program has been tested on Hikey board with octa CA53 CPUs, below
is one example for statistics result, the format mainly follows up
Jesper Dangaard Brouer suggestion.
Jesper reminds to 'get printf to pretty print with thousands separators
use %' and setlocale(LC_NUMERIC, "en_US")', tried three different arm64
GCC toolchains (5.4.0 20160609, 6.2.1 20161016, 6.3.0 20170516) but all
of them cannot support printf flag character %' on arm64 platform, so go
back print number without grouping mode.
CPU states statistics:
state(ms) cstate-0 cstate-1 cstate-2 pstate-0 pstate-1 pstate-2 pstate-3 pstate-4
CPU-0 767 6111 111863 561 31 756 853 190
CPU-1 241 10606 107956 484 125 646 990 85
CPU-2 413 19721 98735 636 84 696 757 89
CPU-3 84 11711 79989 17516 909 4811 5773 341
CPU-4 152 19610 98229 444 53 649 708 1283
CPU-5 185 8781 108697 666 91 671 677 1365
CPU-6 157 21964 95825 581 67 566 684 1284
CPU-7 125 15238 102704 398 20 665 786 1197
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-02-26 04:19:12 +03:00
|
|
|
char filename[256];
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
|
2020-08-23 11:53:34 +03:00
|
|
|
obj = bpf_object__open_file(filename, NULL);
|
|
|
|
if (libbpf_get_error(obj)) {
|
|
|
|
fprintf(stderr, "ERROR: opening BPF object file failed\n");
|
|
|
|
return 0;
|
|
|
|
}
|
samples/bpf: Add program for CPU state statistics
CPU is active when have running tasks on it and CPUFreq governor can
select different operating points (OPP) according to different workload;
we use 'pstate' to present CPU state which have running tasks with one
specific OPP. On the other hand, CPU is idle which only idle task on
it, CPUIdle governor can select one specific idle state to power off
hardware logics; we use 'cstate' to present CPU idle state.
Based on trace events 'cpu_idle' and 'cpu_frequency' we can accomplish
the duration statistics for every state. Every time when CPU enters
into or exits from idle states, the trace event 'cpu_idle' is recorded;
trace event 'cpu_frequency' records the event for CPU OPP changing, so
it's easily to know how long time the CPU stays in the specified OPP,
and the CPU must be not in any idle state.
This patch is to utilize the mentioned trace events for pstate and
cstate statistics. To achieve more accurate profiling data, the program
uses below sequence to insure CPU running/idle time aren't missed:
- Before profiling the user space program wakes up all CPUs for once, so
can avoid to missing account time for CPU staying in idle state for
long time; the program forces to set 'scaling_max_freq' to lowest
frequency and then restore 'scaling_max_freq' to highest frequency,
this can ensure the frequency to be set to lowest frequency and later
after start to run workload the frequency can be easily to be changed
to higher frequency;
- User space program reads map data and update statistics for every 5s,
so this is same with other sample bpf programs for avoiding big
overload introduced by bpf program self;
- When send signal to terminate program, the signal handler wakes up
all CPUs, set lowest frequency and restore highest frequency to
'scaling_max_freq'; this is exactly same with the first step so
avoid to missing account CPU pstate and cstate time during last
stage. Finally it reports the latest statistics.
The program has been tested on Hikey board with octa CA53 CPUs, below
is one example for statistics result, the format mainly follows up
Jesper Dangaard Brouer suggestion.
Jesper reminds to 'get printf to pretty print with thousands separators
use %' and setlocale(LC_NUMERIC, "en_US")', tried three different arm64
GCC toolchains (5.4.0 20160609, 6.2.1 20161016, 6.3.0 20170516) but all
of them cannot support printf flag character %' on arm64 platform, so go
back print number without grouping mode.
CPU states statistics:
state(ms) cstate-0 cstate-1 cstate-2 pstate-0 pstate-1 pstate-2 pstate-3 pstate-4
CPU-0 767 6111 111863 561 31 756 853 190
CPU-1 241 10606 107956 484 125 646 990 85
CPU-2 413 19721 98735 636 84 696 757 89
CPU-3 84 11711 79989 17516 909 4811 5773 341
CPU-4 152 19610 98229 444 53 649 708 1283
CPU-5 185 8781 108697 666 91 671 677 1365
CPU-6 157 21964 95825 581 67 566 684 1284
CPU-7 125 15238 102704 398 20 665 786 1197
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-02-26 04:19:12 +03:00
|
|
|
|
2020-08-23 11:53:34 +03:00
|
|
|
prog = bpf_object__find_program_by_name(obj, "bpf_prog1");
|
|
|
|
if (!prog) {
|
|
|
|
printf("finding a prog in obj file failed\n");
|
|
|
|
goto cleanup;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* load BPF program */
|
|
|
|
if (bpf_object__load(obj)) {
|
|
|
|
fprintf(stderr, "ERROR: loading BPF object file failed\n");
|
|
|
|
goto cleanup;
|
|
|
|
}
|
|
|
|
|
|
|
|
cstate_map_fd = bpf_object__find_map_fd_by_name(obj, "cstate_duration");
|
|
|
|
pstate_map_fd = bpf_object__find_map_fd_by_name(obj, "pstate_duration");
|
|
|
|
if (cstate_map_fd < 0 || pstate_map_fd < 0) {
|
|
|
|
fprintf(stderr, "ERROR: finding a map in obj file failed\n");
|
|
|
|
goto cleanup;
|
|
|
|
}
|
|
|
|
|
|
|
|
link = bpf_program__attach(prog);
|
|
|
|
if (libbpf_get_error(link)) {
|
|
|
|
fprintf(stderr, "ERROR: bpf_program__attach failed\n");
|
|
|
|
link = NULL;
|
|
|
|
goto cleanup;
|
samples/bpf: Add program for CPU state statistics
CPU is active when have running tasks on it and CPUFreq governor can
select different operating points (OPP) according to different workload;
we use 'pstate' to present CPU state which have running tasks with one
specific OPP. On the other hand, CPU is idle which only idle task on
it, CPUIdle governor can select one specific idle state to power off
hardware logics; we use 'cstate' to present CPU idle state.
Based on trace events 'cpu_idle' and 'cpu_frequency' we can accomplish
the duration statistics for every state. Every time when CPU enters
into or exits from idle states, the trace event 'cpu_idle' is recorded;
trace event 'cpu_frequency' records the event for CPU OPP changing, so
it's easily to know how long time the CPU stays in the specified OPP,
and the CPU must be not in any idle state.
This patch is to utilize the mentioned trace events for pstate and
cstate statistics. To achieve more accurate profiling data, the program
uses below sequence to insure CPU running/idle time aren't missed:
- Before profiling the user space program wakes up all CPUs for once, so
can avoid to missing account time for CPU staying in idle state for
long time; the program forces to set 'scaling_max_freq' to lowest
frequency and then restore 'scaling_max_freq' to highest frequency,
this can ensure the frequency to be set to lowest frequency and later
after start to run workload the frequency can be easily to be changed
to higher frequency;
- User space program reads map data and update statistics for every 5s,
so this is same with other sample bpf programs for avoiding big
overload introduced by bpf program self;
- When send signal to terminate program, the signal handler wakes up
all CPUs, set lowest frequency and restore highest frequency to
'scaling_max_freq'; this is exactly same with the first step so
avoid to missing account CPU pstate and cstate time during last
stage. Finally it reports the latest statistics.
The program has been tested on Hikey board with octa CA53 CPUs, below
is one example for statistics result, the format mainly follows up
Jesper Dangaard Brouer suggestion.
Jesper reminds to 'get printf to pretty print with thousands separators
use %' and setlocale(LC_NUMERIC, "en_US")', tried three different arm64
GCC toolchains (5.4.0 20160609, 6.2.1 20161016, 6.3.0 20170516) but all
of them cannot support printf flag character %' on arm64 platform, so go
back print number without grouping mode.
CPU states statistics:
state(ms) cstate-0 cstate-1 cstate-2 pstate-0 pstate-1 pstate-2 pstate-3 pstate-4
CPU-0 767 6111 111863 561 31 756 853 190
CPU-1 241 10606 107956 484 125 646 990 85
CPU-2 413 19721 98735 636 84 696 757 89
CPU-3 84 11711 79989 17516 909 4811 5773 341
CPU-4 152 19610 98229 444 53 649 708 1283
CPU-5 185 8781 108697 666 91 671 677 1365
CPU-6 157 21964 95825 581 67 566 684 1284
CPU-7 125 15238 102704 398 20 665 786 1197
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-02-26 04:19:12 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
ret = cpu_stat_inject_cpu_idle_event();
|
|
|
|
if (ret < 0)
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
ret = cpu_stat_inject_cpu_frequency_event();
|
|
|
|
if (ret < 0)
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
signal(SIGINT, int_exit);
|
|
|
|
signal(SIGTERM, int_exit);
|
|
|
|
|
|
|
|
while (1) {
|
2020-08-23 11:53:34 +03:00
|
|
|
cpu_stat_update(cstate_map_fd, pstate_map_fd);
|
samples/bpf: Add program for CPU state statistics
CPU is active when have running tasks on it and CPUFreq governor can
select different operating points (OPP) according to different workload;
we use 'pstate' to present CPU state which have running tasks with one
specific OPP. On the other hand, CPU is idle which only idle task on
it, CPUIdle governor can select one specific idle state to power off
hardware logics; we use 'cstate' to present CPU idle state.
Based on trace events 'cpu_idle' and 'cpu_frequency' we can accomplish
the duration statistics for every state. Every time when CPU enters
into or exits from idle states, the trace event 'cpu_idle' is recorded;
trace event 'cpu_frequency' records the event for CPU OPP changing, so
it's easily to know how long time the CPU stays in the specified OPP,
and the CPU must be not in any idle state.
This patch is to utilize the mentioned trace events for pstate and
cstate statistics. To achieve more accurate profiling data, the program
uses below sequence to insure CPU running/idle time aren't missed:
- Before profiling the user space program wakes up all CPUs for once, so
can avoid to missing account time for CPU staying in idle state for
long time; the program forces to set 'scaling_max_freq' to lowest
frequency and then restore 'scaling_max_freq' to highest frequency,
this can ensure the frequency to be set to lowest frequency and later
after start to run workload the frequency can be easily to be changed
to higher frequency;
- User space program reads map data and update statistics for every 5s,
so this is same with other sample bpf programs for avoiding big
overload introduced by bpf program self;
- When send signal to terminate program, the signal handler wakes up
all CPUs, set lowest frequency and restore highest frequency to
'scaling_max_freq'; this is exactly same with the first step so
avoid to missing account CPU pstate and cstate time during last
stage. Finally it reports the latest statistics.
The program has been tested on Hikey board with octa CA53 CPUs, below
is one example for statistics result, the format mainly follows up
Jesper Dangaard Brouer suggestion.
Jesper reminds to 'get printf to pretty print with thousands separators
use %' and setlocale(LC_NUMERIC, "en_US")', tried three different arm64
GCC toolchains (5.4.0 20160609, 6.2.1 20161016, 6.3.0 20170516) but all
of them cannot support printf flag character %' on arm64 platform, so go
back print number without grouping mode.
CPU states statistics:
state(ms) cstate-0 cstate-1 cstate-2 pstate-0 pstate-1 pstate-2 pstate-3 pstate-4
CPU-0 767 6111 111863 561 31 756 853 190
CPU-1 241 10606 107956 484 125 646 990 85
CPU-2 413 19721 98735 636 84 696 757 89
CPU-3 84 11711 79989 17516 909 4811 5773 341
CPU-4 152 19610 98229 444 53 649 708 1283
CPU-5 185 8781 108697 666 91 671 677 1365
CPU-6 157 21964 95825 581 67 566 684 1284
CPU-7 125 15238 102704 398 20 665 786 1197
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-02-26 04:19:12 +03:00
|
|
|
cpu_stat_print();
|
|
|
|
sleep(5);
|
|
|
|
}
|
|
|
|
|
2020-08-23 11:53:34 +03:00
|
|
|
cleanup:
|
|
|
|
bpf_link__destroy(link);
|
|
|
|
bpf_object__close(obj);
|
samples/bpf: Add program for CPU state statistics
CPU is active when have running tasks on it and CPUFreq governor can
select different operating points (OPP) according to different workload;
we use 'pstate' to present CPU state which have running tasks with one
specific OPP. On the other hand, CPU is idle which only idle task on
it, CPUIdle governor can select one specific idle state to power off
hardware logics; we use 'cstate' to present CPU idle state.
Based on trace events 'cpu_idle' and 'cpu_frequency' we can accomplish
the duration statistics for every state. Every time when CPU enters
into or exits from idle states, the trace event 'cpu_idle' is recorded;
trace event 'cpu_frequency' records the event for CPU OPP changing, so
it's easily to know how long time the CPU stays in the specified OPP,
and the CPU must be not in any idle state.
This patch is to utilize the mentioned trace events for pstate and
cstate statistics. To achieve more accurate profiling data, the program
uses below sequence to insure CPU running/idle time aren't missed:
- Before profiling the user space program wakes up all CPUs for once, so
can avoid to missing account time for CPU staying in idle state for
long time; the program forces to set 'scaling_max_freq' to lowest
frequency and then restore 'scaling_max_freq' to highest frequency,
this can ensure the frequency to be set to lowest frequency and later
after start to run workload the frequency can be easily to be changed
to higher frequency;
- User space program reads map data and update statistics for every 5s,
so this is same with other sample bpf programs for avoiding big
overload introduced by bpf program self;
- When send signal to terminate program, the signal handler wakes up
all CPUs, set lowest frequency and restore highest frequency to
'scaling_max_freq'; this is exactly same with the first step so
avoid to missing account CPU pstate and cstate time during last
stage. Finally it reports the latest statistics.
The program has been tested on Hikey board with octa CA53 CPUs, below
is one example for statistics result, the format mainly follows up
Jesper Dangaard Brouer suggestion.
Jesper reminds to 'get printf to pretty print with thousands separators
use %' and setlocale(LC_NUMERIC, "en_US")', tried three different arm64
GCC toolchains (5.4.0 20160609, 6.2.1 20161016, 6.3.0 20170516) but all
of them cannot support printf flag character %' on arm64 platform, so go
back print number without grouping mode.
CPU states statistics:
state(ms) cstate-0 cstate-1 cstate-2 pstate-0 pstate-1 pstate-2 pstate-3 pstate-4
CPU-0 767 6111 111863 561 31 756 853 190
CPU-1 241 10606 107956 484 125 646 990 85
CPU-2 413 19721 98735 636 84 696 757 89
CPU-3 84 11711 79989 17516 909 4811 5773 341
CPU-4 152 19610 98229 444 53 649 708 1283
CPU-5 185 8781 108697 666 91 671 677 1365
CPU-6 157 21964 95825 581 67 566 684 1284
CPU-7 125 15238 102704 398 20 665 786 1197
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-02-26 04:19:12 +03:00
|
|
|
return 0;
|
|
|
|
}
|