* [PATCH] cpuidle: Add cpu_idle_miss trace event
@ 2022-07-26 10:24 Kajetan Puchalski
2022-07-26 14:08 ` Steven Rostedt
2022-07-29 14:57 ` Rafael J. Wysocki
0 siblings, 2 replies; 4+ messages in thread
From: Kajetan Puchalski @ 2022-07-26 10:24 UTC (permalink / raw)
To: linux-kernel, linux-pm
Cc: Rafael J. Wysocki, Daniel Lezcano, Steven Rostedt, Ingo Molnar,
lukasz.luba, Dietmar.Eggemann, kajetan.puchalski
Add a trace event for cpuidle to track missed (too deep or too shallow)
wakeups.
After each wakeup, CPUIdle already computes whether the entered state was
optimal, above or below the desired one and updates the relevant
counters. This patch makes it possible to trace those events in addition
to just reading the counters.
The patterns of types and percentages of misses across different
workloads appear to be very consistent. This makes the trace event very
useful for comparing the relative correctness of different CPUIdle
governors for different types of workloads, or for finding the
optimal governor for a given device.
Signed-off-by: Kajetan Puchalski <kajetan.puchalski@arm.com>
---
drivers/cpuidle/cpuidle.c | 6 +++++-
include/trace/events/power.h | 22 ++++++++++++++++++++++
2 files changed, 27 insertions(+), 1 deletion(-)
diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index ef2ea1b12cd8..bf57cab32456 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -8,6 +8,7 @@
* This code is licenced under the GPL.
*/
+#include "linux/percpu-defs.h"
#include <linux/clockchips.h>
#include <linux/kernel.h>
#include <linux/mutex.h>
@@ -278,6 +279,7 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv,
/* Shallower states are enabled, so update. */
dev->states_usage[entered_state].above++;
+ trace_cpu_idle_miss(dev->cpu, entered_state, false);
break;
}
} else if (diff > delay) {
@@ -289,8 +291,10 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv,
* Update if a deeper state would have been a
* better match for the observed idle duration.
*/
- if (diff - delay >= drv->states[i].target_residency_ns)
+ if (diff - delay >= drv->states[i].target_residency_ns) {
dev->states_usage[entered_state].below++;
+ trace_cpu_idle_miss(dev->cpu, entered_state, true);
+ }
break;
}
diff --git a/include/trace/events/power.h b/include/trace/events/power.h
index af5018aa9517..6539f23a5653 100644
--- a/include/trace/events/power.h
+++ b/include/trace/events/power.h
@@ -40,6 +40,28 @@ DEFINE_EVENT(cpu, cpu_idle,
TP_ARGS(state, cpu_id)
);
+TRACE_EVENT(cpu_idle_miss,
+
+ TP_PROTO(unsigned int cpu_id, unsigned int state, bool below),
+
+ TP_ARGS(cpu_id, state, below),
+
+ TP_STRUCT__entry(
+ __field(u32, cpu_id)
+ __field(u32, state)
+ __field(bool, below)
+ ),
+
+ TP_fast_assign(
+ __entry->cpu_id = cpu_id;
+ __entry->state = state;
+ __entry->below = below;
+ ),
+
+ TP_printk("cpu_id=%lu state=%lu type=%s", (unsigned long)__entry->cpu_id,
+ (unsigned long)__entry->state, (__entry->below)?"below":"above")
+);
+
TRACE_EVENT(powernv_throttle,
TP_PROTO(int chip_id, const char *reason, int pmax),
base-commit: e0dccc3b76fb35bb257b4118367a883073d7390e
--
2.37.1
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH] cpuidle: Add cpu_idle_miss trace event
2022-07-26 10:24 [PATCH] cpuidle: Add cpu_idle_miss trace event Kajetan Puchalski
@ 2022-07-26 14:08 ` Steven Rostedt
2022-07-26 15:59 ` Kajetan Puchalski
2022-07-29 14:57 ` Rafael J. Wysocki
1 sibling, 1 reply; 4+ messages in thread
From: Steven Rostedt @ 2022-07-26 14:08 UTC (permalink / raw)
To: Kajetan Puchalski
Cc: linux-kernel, linux-pm, Rafael J. Wysocki, Daniel Lezcano,
Ingo Molnar, lukasz.luba, Dietmar.Eggemann
On Tue, 26 Jul 2022 11:24:04 +0100
Kajetan Puchalski <kajetan.puchalski@arm.com> wrote:
> --- a/include/trace/events/power.h
> +++ b/include/trace/events/power.h
> @@ -40,6 +40,28 @@ DEFINE_EVENT(cpu, cpu_idle,
> TP_ARGS(state, cpu_id)
> );
>
> +TRACE_EVENT(cpu_idle_miss,
> +
> + TP_PROTO(unsigned int cpu_id, unsigned int state, bool below),
> +
> + TP_ARGS(cpu_id, state, below),
> +
> + TP_STRUCT__entry(
> + __field(u32, cpu_id)
> + __field(u32, state)
> + __field(bool, below)
> + ),
> +
> + TP_fast_assign(
> + __entry->cpu_id = cpu_id;
> + __entry->state = state;
> + __entry->below = below;
> + ),
> +
> + TP_printk("cpu_id=%lu state=%lu type=%s", (unsigned long)__entry->cpu_id,
> + (unsigned long)__entry->state, (__entry->below)?"below":"above")
> +);
> +
> TRACE_EVENT(powernv_throttle,
>
> TP_PROTO(int chip_id, const char *reason, int pmax),
For the tracing POV,
Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
-- Steve
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] cpuidle: Add cpu_idle_miss trace event
2022-07-26 14:08 ` Steven Rostedt
@ 2022-07-26 15:59 ` Kajetan Puchalski
0 siblings, 0 replies; 4+ messages in thread
From: Kajetan Puchalski @ 2022-07-26 15:59 UTC (permalink / raw)
To: Steven Rostedt
Cc: linux-kernel, linux-pm, Rafael J. Wysocki, Daniel Lezcano,
Ingo Molnar, lukasz.luba, Dietmar.Eggemann, kajetan.puchalski
On Tue, Jul 26, 2022 at 10:08:16AM -0400, Steven Rostedt wrote:
> On Tue, 26 Jul 2022 11:24:04 +0100
> Kajetan Puchalski <kajetan.puchalski@arm.com> wrote:
> For the tracing POV,
>
> Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Thanks a lot for taking a look!
> -- Steve
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] cpuidle: Add cpu_idle_miss trace event
2022-07-26 10:24 [PATCH] cpuidle: Add cpu_idle_miss trace event Kajetan Puchalski
2022-07-26 14:08 ` Steven Rostedt
@ 2022-07-29 14:57 ` Rafael J. Wysocki
1 sibling, 0 replies; 4+ messages in thread
From: Rafael J. Wysocki @ 2022-07-29 14:57 UTC (permalink / raw)
To: Kajetan Puchalski
Cc: Linux Kernel Mailing List, Linux PM, Rafael J. Wysocki,
Daniel Lezcano, Steven Rostedt, Ingo Molnar, Lukasz Luba,
Dietmar Eggemann
On Tue, Jul 26, 2022 at 12:24 PM Kajetan Puchalski
<kajetan.puchalski@arm.com> wrote:
>
> Add a trace event for cpuidle to track missed (too deep or too shallow)
> wakeups.
>
> After each wakeup, CPUIdle already computes whether the entered state was
> optimal, above or below the desired one and updates the relevant
> counters. This patch makes it possible to trace those events in addition
> to just reading the counters.
>
> The patterns of types and percentages of misses across different
> workloads appear to be very consistent. This makes the trace event very
> useful for comparing the relative correctness of different CPUIdle
> governors for different types of workloads, or for finding the
> optimal governor for a given device.
>
> Signed-off-by: Kajetan Puchalski <kajetan.puchalski@arm.com>
> ---
> drivers/cpuidle/cpuidle.c | 6 +++++-
> include/trace/events/power.h | 22 ++++++++++++++++++++++
> 2 files changed, 27 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
> index ef2ea1b12cd8..bf57cab32456 100644
> --- a/drivers/cpuidle/cpuidle.c
> +++ b/drivers/cpuidle/cpuidle.c
> @@ -8,6 +8,7 @@
> * This code is licenced under the GPL.
> */
>
> +#include "linux/percpu-defs.h"
> #include <linux/clockchips.h>
> #include <linux/kernel.h>
> #include <linux/mutex.h>
> @@ -278,6 +279,7 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv,
>
> /* Shallower states are enabled, so update. */
> dev->states_usage[entered_state].above++;
> + trace_cpu_idle_miss(dev->cpu, entered_state, false);
> break;
> }
> } else if (diff > delay) {
> @@ -289,8 +291,10 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv,
> * Update if a deeper state would have been a
> * better match for the observed idle duration.
> */
> - if (diff - delay >= drv->states[i].target_residency_ns)
> + if (diff - delay >= drv->states[i].target_residency_ns) {
> dev->states_usage[entered_state].below++;
> + trace_cpu_idle_miss(dev->cpu, entered_state, true);
> + }
>
> break;
> }
> diff --git a/include/trace/events/power.h b/include/trace/events/power.h
> index af5018aa9517..6539f23a5653 100644
> --- a/include/trace/events/power.h
> +++ b/include/trace/events/power.h
> @@ -40,6 +40,28 @@ DEFINE_EVENT(cpu, cpu_idle,
> TP_ARGS(state, cpu_id)
> );
>
> +TRACE_EVENT(cpu_idle_miss,
> +
> + TP_PROTO(unsigned int cpu_id, unsigned int state, bool below),
> +
> + TP_ARGS(cpu_id, state, below),
> +
> + TP_STRUCT__entry(
> + __field(u32, cpu_id)
> + __field(u32, state)
> + __field(bool, below)
> + ),
> +
> + TP_fast_assign(
> + __entry->cpu_id = cpu_id;
> + __entry->state = state;
> + __entry->below = below;
> + ),
> +
> + TP_printk("cpu_id=%lu state=%lu type=%s", (unsigned long)__entry->cpu_id,
> + (unsigned long)__entry->state, (__entry->below)?"below":"above")
> +);
> +
> TRACE_EVENT(powernv_throttle,
>
> TP_PROTO(int chip_id, const char *reason, int pmax),
>
> base-commit: e0dccc3b76fb35bb257b4118367a883073d7390e
> --
Applied as 5.20 material with the R-by tag from Steve, thanks!
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2022-07-29 14:58 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-07-26 10:24 [PATCH] cpuidle: Add cpu_idle_miss trace event Kajetan Puchalski
2022-07-26 14:08 ` Steven Rostedt
2022-07-26 15:59 ` Kajetan Puchalski
2022-07-29 14:57 ` Rafael J. Wysocki
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.