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

@@ -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);

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;
 }
 

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 */

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),

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;

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);

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)

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);