tracing: add "power-tracer": C/P state tracer to help power optimization

Impact: new "power-tracer" ftrace plugin

This patch adds a C/P-state ftrace plugin that will generate
detailed statistics about the C/P-states that are being used,
so that we can look at detailed decisions that the C/P-state
code is making, rather than the too high level "average"
that we have today.

An example way of using this is:

 mount -t debugfs none /sys/kernel/debug
 echo cstate > /sys/kernel/debug/tracing/current_tracer
 echo 1 > /sys/kernel/debug/tracing/tracing_enabled
 sleep 1
 echo 0 > /sys/kernel/debug/tracing/tracing_enabled
 cat /sys/kernel/debug/tracing/trace | perl scripts/trace/cstate.pl > out.svg

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

2 files changed, 20 insertions, 0 deletions

diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index 8e48c5d4467d..88ea02dcb622 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -33,6 +33,7 @@
 #include <linux/cpufreq.h>
 #include <linux/compiler.h>
 #include <linux/dmi.h>
+#include <linux/ftrace.h>
 
 #include <linux/acpi.h>
 #include <acpi/processor.h>
@@ -391,6 +392,7 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
         unsigned int next_perf_state = 0; /* Index into perf table */
         unsigned int i;
         int result = 0;
+        struct power_trace it;
 
         dprintk("acpi_cpufreq_target %d (%d)\n", target_freq, policy->cpu);
 
@@ -427,6 +429,8 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
                 }
         }
 
+        trace_power_mark(&it, POWER_PSTATE, next_perf_state);
+
         switch (data->cpu_feature) {
         case SYSTEM_INTEL_MSR_CAPABLE:
                 cmd.type = SYSTEM_INTEL_MSR_CAPABLE;
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index c622772744d8..c27af49a4ede 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -7,6 +7,7 @@
 #include <linux/module.h>
 #include <linux/pm.h>
 #include <linux/clockchips.h>
+#include <linux/ftrace.h>
 #include <asm/system.h>
 
 unsigned long idle_halt;
@@ -100,6 +101,9 @@ static inline int hlt_use_halt(void)
 void default_idle(void)
 {
         if (hlt_use_halt()) {
+                struct power_trace it;
+
+                trace_power_start(&it, POWER_CSTATE, 1);
                 current_thread_info()->status &= ~TS_POLLING;
                 /*
                  * TS_POLLING-cleared state must be visible before we
@@ -112,6 +116,7 @@ void default_idle(void)
                 else
                         local_irq_enable();
                 current_thread_info()->status |= TS_POLLING;
+                trace_power_end(&it);
         } else {
                 local_irq_enable();
                 /* loop is done by the caller */
@@ -154,24 +159,31 @@ EXPORT_SYMBOL_GPL(cpu_idle_wait);
  */
 void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
 {
+        struct power_trace it;
+
+        trace_power_start(&it, POWER_CSTATE, (ax>>4)+1);
         if (!need_resched()) {
                 __monitor((void *)&current_thread_info()->flags, 0, 0);
                 smp_mb();
                 if (!need_resched())
                         __mwait(ax, cx);
         }
+        trace_power_end(&it);
 }
 
 /* Default MONITOR/MWAIT with no hints, used for default C1 state */
 static void mwait_idle(void)
 {
+        struct power_trace it;
         if (!need_resched()) {
+                trace_power_start(&it, POWER_CSTATE, 1);
                 __monitor((void *)&current_thread_info()->flags, 0, 0);
                 smp_mb();
                 if (!need_resched())
                         __sti_mwait(0, 0);
                 else
                         local_irq_enable();
+                trace_power_end(&it);
         } else
                 local_irq_enable();
 }
@@ -183,9 +195,13 @@ static void mwait_idle(void)
  */
 static void poll_idle(void)
 {
+        struct power_trace it;
+
+        trace_power_start(&it, POWER_CSTATE, 0);
         local_irq_enable();
         while (!need_resched())
                 cpu_relax();
+        trace_power_end(&it);
 }
 
 /*