From bb3632c6101b2fad07e6246721466b984b1e0e9d Mon Sep 17 00:00:00 2001
From: Todd E Brandt <todd.e.brandt@linux.intel.com>
Date: Fri, 6 Jun 2014 05:40:17 -0700
Subject: [PATCH 1/2] PM / sleep: trace events for suspend/resume
Adds trace events that give finer resolution into suspend/resume. These
events are graphed in the timelines generated by the analyze_suspend.py
script. They represent large areas of time consumed that are typical to
suspend and resume.
The event is triggered by calling the function "trace_suspend_resume"
with three arguments: a string (the name of the event to be displayed
in the timeline), an integer (case specific number, such as the power
state or cpu number), and a boolean (where true is used to denote the start
of the timeline event, and false to denote the end).
The suspend_resume trace event reproduces the data that the machine_suspend
trace event did, so the latter has been removed.
Signed-off-by: Todd Brandt <todd.e.brandt@intel.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
drivers/acpi/sleep.c | 3 +++
drivers/base/power/main.c | 16 ++++++++++++++
drivers/base/syscore.c | 5 +++++
include/trace/events/power.h | 42 +++++++++++++++++++++---------------
kernel/cpu.c | 5 +++++
kernel/power/hibernate.c | 3 +++
kernel/power/process.c | 3 +++
kernel/power/suspend.c | 14 ++++++++++--
8 files changed, 72 insertions(+), 19 deletions(-)
diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c
index c11e3795431b..b3e3cc73ba79 100644
--- a/drivers/acpi/sleep.c
+++ b/drivers/acpi/sleep.c
@@ -19,6 +19,7 @@
#include <linux/acpi.h>
#include <linux/module.h>
#include <asm/io.h>
+#include <trace/events/power.h>
#include "internal.h"
#include "sleep.h"
@@ -501,6 +502,7 @@ static int acpi_suspend_enter(suspend_state_t pm_state)
ACPI_FLUSH_CPU_CACHE();
+ trace_suspend_resume(TPS("acpi_suspend"), acpi_state, true);
switch (acpi_state) {
case ACPI_STATE_S1:
barrier();
@@ -516,6 +518,7 @@ static int acpi_suspend_enter(suspend_state_t pm_state)
pr_info(PREFIX "Low-level resume complete\n");
break;
}
+ trace_suspend_resume(TPS("acpi_suspend"), acpi_state, false);
/* This violates the spec but is required for bug compatibility. */
acpi_write_bit_register(ACPI_BITREG_SCI_ENABLE, 1);
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index 343ffad59377..de3fe4fe350a 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -545,6 +545,7 @@ static void dpm_resume_noirq(pm_message_t state)
struct device *dev;
ktime_t starttime = ktime_get();
+ trace_suspend_resume(TPS("dpm_resume_noirq"), state.event, true);
mutex_lock(&dpm_list_mtx);
pm_transition = state;
@@ -587,6 +588,7 @@ static void dpm_resume_noirq(pm_message_t state)
dpm_show_time(starttime, state, "noirq");
resume_device_irqs();
cpuidle_resume();
+ trace_suspend_resume(TPS("dpm_resume_noirq"), state.event, false);
}
/**
@@ -664,6 +666,7 @@ static void dpm_resume_early(pm_message_t state)
struct device *dev;
ktime_t starttime = ktime_get();
+ trace_suspend_resume(TPS("dpm_resume_early"), state.event, true);
mutex_lock(&dpm_list_mtx);
pm_transition = state;
@@ -703,6 +706,7 @@ static void dpm_resume_early(pm_message_t state)
mutex_unlock(&dpm_list_mtx);
async_synchronize_full();
dpm_show_time(starttime, state, "early");
+ trace_suspend_resume(TPS("dpm_resume_early"), state.event, false);
}
/**
@@ -834,6 +838,7 @@ void dpm_resume(pm_message_t state)
struct device *dev;
ktime_t starttime = ktime_get();
+ trace_suspend_resume(TPS("dpm_resume"), state.event, true);
might_sleep();
mutex_lock(&dpm_list_mtx);
@@ -875,6 +880,7 @@ void dpm_resume(pm_message_t state)
dpm_show_time(starttime, state, NULL);
cpufreq_resume();
+ trace_suspend_resume(TPS("dpm_resume"), state.event, false);
}
/**
@@ -932,6 +938,7 @@ void dpm_complete(pm_message_t state)
{
struct list_head list;
+ trace_suspend_resume(TPS("dpm_complete"), state.event, true);
might_sleep();
INIT_LIST_HEAD(&list);
@@ -951,6 +958,7 @@ void dpm_complete(pm_message_t state)
}
list_splice(&list, &dpm_list);
mutex_unlock(&dpm_list_mtx);
+ trace_suspend_resume(TPS("dpm_complete"), state.event, false);
}
/**
@@ -1086,6 +1094,7 @@ static int dpm_suspend_noirq(pm_message_t state)
ktime_t starttime = ktime_get();
int error = 0;
+ trace_suspend_resume(TPS("dpm_suspend_noirq"), state.event, true);
cpuidle_pause();
suspend_device_irqs();
mutex_lock(&dpm_list_mtx);
@@ -1126,6 +1135,7 @@ static int dpm_suspend_noirq(pm_message_t state)
} else {
dpm_show_time(starttime, state, "noirq");
}
+ trace_suspend_resume(TPS("dpm_suspend_noirq"), state.event, false);
return error;
}
@@ -1222,6 +1232,7 @@ static int dpm_suspend_late(pm_message_t state)
ktime_t starttime = ktime_get();
int error = 0;
+ trace_suspend_resume(TPS("dpm_suspend_late"), state.event, true);
mutex_lock(&dpm_list_mtx);
pm_transition = state;
async_error = 0;
@@ -1257,6 +1268,7 @@ static int dpm_suspend_late(pm_message_t state)
} else {
dpm_show_time(starttime, state, "late");
}
+ trace_suspend_resume(TPS("dpm_suspend_late"), state.event, false);
return error;
}
@@ -1461,6 +1473,7 @@ int dpm_suspend(pm_message_t state)
ktime_t starttime = ktime_get();
int error = 0;
+ trace_suspend_resume(TPS("dpm_suspend"), state.event, true);
might_sleep();
cpufreq_suspend();
@@ -1498,6 +1511,7 @@ int dpm_suspend(pm_message_t state)
dpm_save_failed_step(SUSPEND_SUSPEND);
} else
dpm_show_time(starttime, state, NULL);
+ trace_suspend_resume(TPS("dpm_suspend"), state.event, false);
return error;
}
@@ -1582,6 +1596,7 @@ int dpm_prepare(pm_message_t state)
{
int error = 0;
+ trace_suspend_resume(TPS("dpm_prepare"), state.event, true);
might_sleep();
mutex_lock(&dpm_list_mtx);
@@ -1612,6 +1627,7 @@ int dpm_prepare(pm_message_t state)
put_device(dev);
}
mutex_unlock(&dpm_list_mtx);
+ trace_suspend_resume(TPS("dpm_prepare"), state.event, false);
return error;
}
diff --git a/drivers/base/syscore.c b/drivers/base/syscore.c
index e8d11b6630ee..dbb8350ea8dc 100644
--- a/drivers/base/syscore.c
+++ b/drivers/base/syscore.c
@@ -10,6 +10,7 @@
#include <linux/mutex.h>
#include <linux/module.h>
#include <linux/interrupt.h>
+#include <trace/events/power.h>
static LIST_HEAD(syscore_ops_list);
static DEFINE_MUTEX(syscore_ops_lock);
@@ -49,6 +50,7 @@ int syscore_suspend(void)
struct syscore_ops *ops;
int ret = 0;
+ trace_suspend_resume(TPS("syscore_suspend"), 0, true);
pr_debug("Checking wakeup interrupts\n");
/* Return error code if there are any wakeup interrupts pending. */
@@ -70,6 +72,7 @@ int syscore_suspend(void)
"Interrupts enabled after %pF\n", ops->suspend);
}
+ trace_suspend_resume(TPS("syscore_suspend"), 0, false);
return 0;
err_out:
@@ -92,6 +95,7 @@ void syscore_resume(void)
{
struct syscore_ops *ops;
+ trace_suspend_resume(TPS("syscore_resume"), 0, true);
WARN_ONCE(!irqs_disabled(),
"Interrupts enabled before system core resume.\n");
@@ -103,6 +107,7 @@ void syscore_resume(void)
WARN_ONCE(!irqs_disabled(),
"Interrupts enabled after %pF\n", ops->resume);
}
+ trace_suspend_resume(TPS("syscore_resume"), 0, false);
}
EXPORT_SYMBOL_GPL(syscore_resume);
#endif /* CONFIG_PM_SLEEP */
diff --git a/include/trace/events/power.h b/include/trace/events/power.h
index 9a7e08d61258..f88c8573e66c 100644
--- a/include/trace/events/power.h
+++ b/include/trace/events/power.h
@@ -7,6 +7,9 @@
#include <linux/ktime.h>
#include <linux/pm_qos.h>
#include <linux/tracepoint.h>
+#include <linux/ftrace_event.h>
+
+#define TPS(x) tracepoint_string(x)
DECLARE_EVENT_CLASS(cpu,
@@ -97,23 +100,6 @@ DEFINE_EVENT(cpu, cpu_frequency,
TP_ARGS(frequency, cpu_id)
);
-TRACE_EVENT(machine_suspend,
-
- TP_PROTO(unsigned int state),
-
- TP_ARGS(state),
-
- TP_STRUCT__entry(
- __field( u32, state )
- ),
-
- TP_fast_assign(
- __entry->state = state;
- ),
-
- TP_printk("state=%lu", (unsigned long)__entry->state)
-);
-
TRACE_EVENT(device_pm_report_time,
TP_PROTO(struct device *dev, const char *pm_ops, s64 ops_time,
@@ -151,6 +137,28 @@ TRACE_EVENT(device_pm_report_time,
__entry->ops_time, __entry->error)
);
+TRACE_EVENT(suspend_resume,
+
+ TP_PROTO(const char *action, int val, bool start),
+
+ TP_ARGS(action, val, start),
+
+ TP_STRUCT__entry(
+ __field(const char *, action)
+ __field(int, val)
+ __field(bool, start)
+ ),
+
+ TP_fast_assign(
+ __entry->action = action;
+ __entry->val = val;
+ __entry->start = start;
+ ),
+
+ TP_printk("%s[%u] %s", __entry->action, (unsigned int)__entry->val,
+ (__entry->start)?"begin":"end")
+);
+
DECLARE_EVENT_CLASS(wakeup_source,
TP_PROTO(const char *name, unsigned int state),
diff --git a/kernel/cpu.c b/kernel/cpu.c
index a9e710eef0e2..76deba01322a 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -20,6 +20,7 @@
#include <linux/gfp.h>
#include <linux/suspend.h>
#include <linux/lockdep.h>
+#include <trace/events/power.h>
#include "smpboot.h"
@@ -522,7 +523,9 @@ int disable_nonboot_cpus(void)
for_each_online_cpu(cpu) {
if (cpu == first_cpu)
continue;
+ trace_suspend_resume(TPS("CPU_OFF"), cpu, true);
error = _cpu_down(cpu, 1);
+ trace_suspend_resume(TPS("CPU_OFF"), cpu, false);
if (!error)
cpumask_set_cpu(cpu, frozen_cpus);
else {
@@ -566,7 +569,9 @@ void __ref enable_nonboot_cpus(void)
arch_enable_nonboot_cpus_begin();
for_each_cpu(cpu, frozen_cpus) {
+ trace_suspend_resume(TPS("CPU_ON"), cpu, true);
error = _cpu_up(cpu, 1);
+ trace_suspend_resume(TPS("CPU_ON"), cpu, false);
if (!error) {
printk(KERN_INFO "CPU%d is up\n", cpu);
continue;
diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index df88d55dc436..49e0a20fd010 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -28,6 +28,7 @@
#include <linux/syscore_ops.h>
#include <linux/ctype.h>
#include <linux/genhd.h>
+#include <trace/events/power.h>
#include "power.h"
@@ -292,7 +293,9 @@ static int create_image(int platform_mode)
in_suspend = 1;
save_processor_state();
+ trace_suspend_resume(TPS("machine_suspend"), PM_EVENT_HIBERNATE, true);
error = swsusp_arch_suspend();
+ trace_suspend_resume(TPS("machine_suspend"), PM_EVENT_HIBERNATE, false);
if (error)
printk(KERN_ERR "PM: Error %d creating hibernation image\n",
error);
diff --git a/kernel/power/process.c b/kernel/power/process.c
index 06ec8869dbf1..0ca8d83e2369 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -17,6 +17,7 @@
#include <linux/delay.h>
#include <linux/workqueue.h>
#include <linux/kmod.h>
+#include <trace/events/power.h>
/*
* Timeout for stopping processes
@@ -175,6 +176,7 @@ void thaw_processes(void)
struct task_struct *g, *p;
struct task_struct *curr = current;
+ trace_suspend_resume(TPS("thaw_processes"), 0, true);
if (pm_freezing)
atomic_dec(&system_freezing_cnt);
pm_freezing = false;
@@ -201,6 +203,7 @@ void thaw_processes(void)
schedule();
printk("done.\n");
+ trace_suspend_resume(TPS("thaw_processes"), 0, false);
}
void thaw_kernel_threads(void)
diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c
index 963e6d0f050b..4dd8822f732a 100644
--- a/kernel/power/suspend.c
+++ b/kernel/power/suspend.c
@@ -177,7 +177,9 @@ static int suspend_prepare(suspend_state_t state)
if (error)
goto Finish;
+ trace_suspend_resume(TPS("freeze_processes"), 0, true);
error = suspend_freeze_processes();
+ trace_suspend_resume(TPS("freeze_processes"), 0, false);
if (!error)
return 0;
@@ -240,7 +242,9 @@ static int suspend_enter(suspend_state_t state, bool *wakeup)
* all the devices are suspended.
*/
if (state == PM_SUSPEND_FREEZE) {
+ trace_suspend_resume(TPS("machine_suspend"), state, true);
freeze_enter();
+ trace_suspend_resume(TPS("machine_suspend"), state, false);
goto Platform_wake;
}
@@ -256,7 +260,11 @@ static int suspend_enter(suspend_state_t state, bool *wakeup)
if (!error) {
*wakeup = pm_wakeup_pending();
if (!(suspend_test(TEST_CORE) || *wakeup)) {
+ trace_suspend_resume(TPS("machine_suspend"),
+ state, true);
error = suspend_ops->enter(state);
+ trace_suspend_resume(TPS("machine_suspend"),
+ state, false);
events_check_enabled = false;
}
syscore_resume();
@@ -294,7 +302,6 @@ int suspend_devices_and_enter(suspend_state_t state)
if (need_suspend_ops(state) && !suspend_ops)
return -ENOSYS;
- trace_machine_suspend(state);
if (need_suspend_ops(state) && suspend_ops->begin) {
error = suspend_ops->begin(state);
if (error)
@@ -331,7 +338,6 @@ int suspend_devices_and_enter(suspend_state_t state)
else if (state == PM_SUSPEND_FREEZE && freeze_ops->end)
freeze_ops->end();
- trace_machine_suspend(PWR_EVENT_EXIT);
return error;
Recover_platform:
@@ -365,6 +371,7 @@ static int enter_state(suspend_state_t state)
{
int error;
+ trace_suspend_resume(TPS("suspend_enter"), state, true);
if (state == PM_SUSPEND_FREEZE) {
#ifdef CONFIG_PM_DEBUG
if (pm_test_level != TEST_NONE && pm_test_level <= TEST_CPUS) {
@@ -382,9 +389,11 @@ static int enter_state(suspend_state_t state)
if (state == PM_SUSPEND_FREEZE)
freeze_begin();
+ trace_suspend_resume(TPS("sync_filesystems"), 0, true);
printk(KERN_INFO "PM: Syncing filesystems ... ");
sys_sync();
printk("done.\n");
+ trace_suspend_resume(TPS("sync_filesystems"), 0, false);
pr_debug("PM: Preparing system for %s sleep\n", pm_states[state].label);
error = suspend_prepare(state);
@@ -394,6 +403,7 @@ static int enter_state(suspend_state_t state)
if (suspend_test(TEST_FREEZER))
goto Finish;
+ trace_suspend_resume(TPS("suspend_enter"), state, false);
pr_debug("PM: Entering %s sleep\n", pm_states[state].label);
pm_restrict_gfp_mask();
error = suspend_devices_and_enter(state);
From e8bca479c3f269ebb3a3acea5ef63314bb677060 Mon Sep 17 00:00:00 2001
From: Todd E Brandt <todd.e.brandt@linux.intel.com>
Date: Tue, 10 Jun 2014 07:31:22 -0700
Subject: [PATCH 2/2] PM / sleep: trace events for device PM callbacks
Adds two trace events which supply the same info that initcall_debug
provides, but via ftrace instead of dmesg. The existing initcall_debug
calls require the pm_print_times_enabled var to be set (either via
sysfs or via the kernel cmd line). The new trace events provide all the
same info as the initcall_debug prints but with less overhead, and also
with coverage of device prepare and complete device callbacks.
These events replace the device_pm_report_time event (which has been
removed). device_pm_callback_start is called first and provides the device
and callback info. device_pm_callback_end is called after with the
device name and error info. The time and pid are gathered from the trace
data headers.
Signed-off-by: Todd Brandt <todd.e.brandt@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
drivers/base/power/main.c | 14 ++++++---
include/trace/events/power.h | 60 +++++++++++++++++++++++++-----------
2 files changed, 52 insertions(+), 22 deletions(-)
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index de3fe4fe350a..bf412961a934 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -214,9 +214,6 @@ static void initcall_debug_report(struct device *dev, ktime_t calltime,
pr_info("call %s+ returned %d after %Ld usecs\n", dev_name(dev),
error, (unsigned long long)nsecs >> 10);
}
-
- trace_device_pm_report_time(dev, info, nsecs, pm_verb(state.event),
- error);
}
/**
@@ -387,7 +384,9 @@ static int dpm_run_callback(pm_callback_t cb, struct device *dev,
calltime = initcall_debug_start(dev);
pm_dev_dbg(dev, state, info);
+ trace_device_pm_callback_start(dev, info, state.event);
error = cb(dev);
+ trace_device_pm_callback_end(dev, error);
suspend_report_result(cb, error);
initcall_debug_report(dev, calltime, error, state, info);
@@ -919,7 +918,9 @@ static void device_complete(struct device *dev, pm_message_t state)
if (callback) {
pm_dev_dbg(dev, state, info);
+ trace_device_pm_callback_start(dev, info, state.event);
callback(dev);
+ trace_device_pm_callback_end(dev, 0);
}
device_unlock(dev);
@@ -1307,7 +1308,9 @@ static int legacy_suspend(struct device *dev, pm_message_t state,
calltime = initcall_debug_start(dev);
+ trace_device_pm_callback_start(dev, info, state.event);
error = cb(dev, state);
+ trace_device_pm_callback_end(dev, error);
suspend_report_result(cb, error);
initcall_debug_report(dev, calltime, error, state, info);
@@ -1563,8 +1566,11 @@ static int device_prepare(struct device *dev, pm_message_t state)
callback = dev->driver->pm->prepare;
}
- if (callback)
+ if (callback) {
+ trace_device_pm_callback_start(dev, info, state.event);
ret = callback(dev);
+ trace_device_pm_callback_end(dev, ret);
+ }
device_unlock(dev);
diff --git a/include/trace/events/power.h b/include/trace/events/power.h
index f88c8573e66c..d19840b0cac8 100644
--- a/include/trace/events/power.h
+++ b/include/trace/events/power.h
@@ -93,6 +93,17 @@ TRACE_EVENT(pstate_sample,
#define PWR_EVENT_EXIT -1
#endif
+#define pm_verb_symbolic(event) \
+ __print_symbolic(event, \
+ { PM_EVENT_SUSPEND, "suspend" }, \
+ { PM_EVENT_RESUME, "resume" }, \
+ { PM_EVENT_FREEZE, "freeze" }, \
+ { PM_EVENT_QUIESCE, "quiesce" }, \
+ { PM_EVENT_HIBERNATE, "hibernate" }, \
+ { PM_EVENT_THAW, "thaw" }, \
+ { PM_EVENT_RESTORE, "restore" }, \
+ { PM_EVENT_RECOVER, "recover" })
+
DEFINE_EVENT(cpu, cpu_frequency,
TP_PROTO(unsigned int frequency, unsigned int cpu_id),
@@ -100,41 +111,54 @@ DEFINE_EVENT(cpu, cpu_frequency,
TP_ARGS(frequency, cpu_id)
);
-TRACE_EVENT(device_pm_report_time,
+TRACE_EVENT(device_pm_callback_start,
- TP_PROTO(struct device *dev, const char *pm_ops, s64 ops_time,
- char *pm_event_str, int error),
+ TP_PROTO(struct device *dev, const char *pm_ops, int event),
- TP_ARGS(dev, pm_ops, ops_time, pm_event_str, error),
+ TP_ARGS(dev, pm_ops, event),
TP_STRUCT__entry(
__string(device, dev_name(dev))
__string(driver, dev_driver_string(dev))
__string(parent, dev->parent ? dev_name(dev->parent) : "none")
__string(pm_ops, pm_ops ? pm_ops : "none ")
- __string(pm_event_str, pm_event_str)
- __field(s64, ops_time)
+ __field(int, event)
+ ),
+
+ TP_fast_assign(
+ __assign_str(device, dev_name(dev));
+ __assign_str(driver, dev_driver_string(dev));
+ __assign_str(parent,
+ dev->parent ? dev_name(dev->parent) : "none");
+ __assign_str(pm_ops, pm_ops ? pm_ops : "none ");
+ __entry->event = event;
+ ),
+
+ TP_printk("%s %s, parent: %s, %s[%s]", __get_str(driver),
+ __get_str(device), __get_str(parent), __get_str(pm_ops),
+ pm_verb_symbolic(__entry->event))
+);
+
+TRACE_EVENT(device_pm_callback_end,
+
+ TP_PROTO(struct device *dev, int error),
+
+ TP_ARGS(dev, error),
+
+ TP_STRUCT__entry(
+ __string(device, dev_name(dev))
+ __string(driver, dev_driver_string(dev))
__field(int, error)
),
TP_fast_assign(
- const char *tmp = dev->parent ? dev_name(dev->parent) : "none";
- const char *tmp_i = pm_ops ? pm_ops : "none ";
-
__assign_str(device, dev_name(dev));
__assign_str(driver, dev_driver_string(dev));
- __assign_str(parent, tmp);
- __assign_str(pm_ops, tmp_i);
- __assign_str(pm_event_str, pm_event_str);
- __entry->ops_time = ops_time;
__entry->error = error;
),
- /* ops_str has an extra space at the end */
- TP_printk("%s %s parent=%s state=%s ops=%snsecs=%lld err=%d",
- __get_str(driver), __get_str(device), __get_str(parent),
- __get_str(pm_event_str), __get_str(pm_ops),
- __entry->ops_time, __entry->error)
+ TP_printk("%s %s, err=%d",
+ __get_str(driver), __get_str(device), __entry->error)
);
TRACE_EVENT(suspend_resume,