[v3,7/7] drivers: devfreq: add tracing for scheduling work
diff mbox series

Message ID 1550010238-24002-8-git-send-email-l.luba@partner.samsung.com
State Not Applicable
Headers show
Series
  • drivers: devfreq: fix and optimize workqueue mechanism
Related show

Commit Message

Lukasz Luba Feb. 12, 2019, 10:23 p.m. UTC
This patch add basic tracing of the devfreq workqueue and delayed work.
It aims to capture changes of the polling intervals and device state.

Signed-off-by: Lukasz Luba <l.luba@partner.samsung.com>
---
 drivers/devfreq/devfreq.c | 8 ++++++++
 1 file changed, 8 insertions(+)

Comments

Chanwoo Choi Feb. 14, 2019, 4:57 a.m. UTC | #1
Hi Lukasz,

On 19. 2. 13. 오전 7:23, Lukasz Luba wrote:
> This patch add basic tracing of the devfreq workqueue and delayed work.
> It aims to capture changes of the polling intervals and device state.
> 
> Signed-off-by: Lukasz Luba <l.luba@partner.samsung.com>
> ---
>  drivers/devfreq/devfreq.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c
> index 29e99ce..c1d0d8c 100644
> --- a/drivers/devfreq/devfreq.c
> +++ b/drivers/devfreq/devfreq.c
> @@ -29,6 +29,9 @@
>  #include <linux/of.h>
>  #include "governor.h"
>  
> +#define CREATE_TRACE_POINTS
> +#include <trace/events/devfreq.h>
> +
>  /* The ~30% load threshold used for load calculation (due to fixed point
>   * arithmetic) */
>  #define LOAD_THRESHOLD_IN_DEVICE_USAGE (300)
> @@ -418,6 +421,7 @@ static void devfreq_monitor(struct work_struct *work)
>  	struct devfreq *devfreq = container_of(work,
>  					struct devfreq, work.work);
>  	unsigned int polling_ms;
> +	const char *df_name = dev_name(&devfreq->dev);

nit: You can use 'dev_name(&devfreq->dev)' directly
without defining the separate df_name.

>  
>  	mutex_lock(&devfreq->lock);
>  	polling_ms = devfreq_get_polling_delay(devfreq);
> @@ -429,6 +433,10 @@ static void devfreq_monitor(struct work_struct *work)
>  	schedule_delayed_work(&devfreq->work,
>  			      msecs_to_jiffies(polling_ms));
>  	mutex_unlock(&devfreq->lock);
> +
> +	trace_devfreq_monitor(df_name, devfreq->previous_freq, polling_ms,
> +			      devfreq->last_status.busy_time,
> +			      devfreq->last_status.total_time);

Regardless of type of work,
I think that trace point is necessary for devfreq framework.

Reviewed-by: Chanwoo Choi <cw00.choi@samsung.com>

>  }
>  
>  /**
>

Patch
diff mbox series

diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c
index 29e99ce..c1d0d8c 100644
--- a/drivers/devfreq/devfreq.c
+++ b/drivers/devfreq/devfreq.c
@@ -29,6 +29,9 @@ 
 #include <linux/of.h>
 #include "governor.h"
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/devfreq.h>
+
 /* The ~30% load threshold used for load calculation (due to fixed point
  * arithmetic) */
 #define LOAD_THRESHOLD_IN_DEVICE_USAGE (300)
@@ -418,6 +421,7 @@  static void devfreq_monitor(struct work_struct *work)
 	struct devfreq *devfreq = container_of(work,
 					struct devfreq, work.work);
 	unsigned int polling_ms;
+	const char *df_name = dev_name(&devfreq->dev);
 
 	mutex_lock(&devfreq->lock);
 	polling_ms = devfreq_get_polling_delay(devfreq);
@@ -429,6 +433,10 @@  static void devfreq_monitor(struct work_struct *work)
 	schedule_delayed_work(&devfreq->work,
 			      msecs_to_jiffies(polling_ms));
 	mutex_unlock(&devfreq->lock);
+
+	trace_devfreq_monitor(df_name, devfreq->previous_freq, polling_ms,
+			      devfreq->last_status.busy_time,
+			      devfreq->last_status.total_time);
 }
 
 /**