[PATCH 2/5 RESEND] thermal: exynos: Miscellaneous fixes to support falling threshold interrupt

[PATCH 2/5 RESEND] thermal: exynos: Miscellaneous fixes to support falling threshold interrupt

[Amit Daniel Kachhap]

Below fixes are done to support falling threshold interrupt,
* Falling interrupt status macro corrected according to exynos5 data sheet.
* The get trend function modified to calculate trip temperature correctly.
* The clearing of interrupt status in the isr is now done after handling
  the event that caused the interrupt.

Signed-off-by: Amit Daniel Kachhap <amit.daniel@samsung.com>
---
 drivers/thermal/exynos_thermal.c |   19 ++++++++++++-------
 1 files changed, 12 insertions(+), 7 deletions(-)

diff --git a/drivers/thermal/exynos_thermal.c b/drivers/thermal/exynos_thermal.c
index dcd13f7..0d17d41 100644
--- a/drivers/thermal/exynos_thermal.c
+++ b/drivers/thermal/exynos_thermal.c
@@ -82,7 +82,7 @@
 
 #define EXYNOS_TRIMINFO_RELOAD		0x1
 #define EXYNOS_TMU_CLEAR_RISE_INT	0x111
-#define EXYNOS_TMU_CLEAR_FALL_INT	(0x111 << 16)
+#define EXYNOS_TMU_CLEAR_FALL_INT	(0x111 << 12)
 #define EXYNOS_MUX_ADDR_VALUE		6
 #define EXYNOS_MUX_ADDR_SHIFT		20
 #define EXYNOS_TMU_TRIP_MODE_SHIFT	13
@@ -373,12 +373,19 @@ static int exynos_get_temp(struct thermal_zone_device *thermal,
 static int exynos_get_trend(struct thermal_zone_device *thermal,
 			int trip, enum thermal_trend *trend)
 {
-	if (thermal->temperature >= trip)
+	int ret = 0;
+	unsigned long trip_temp;
+
+	ret = exynos_get_trip_temp(thermal, trip, &trip_temp);
+	if (ret < 0)
+		return ret;
+
+	if (thermal->temperature >= trip_temp)
 		*trend = THERMAL_TREND_RAISING;
 	else
 		*trend = THERMAL_TREND_DROPPING;
 
-	return 0;
+	return ret;
 }
 /* Operation callback functions for thermal zone */
 static struct thermal_zone_device_ops const exynos_dev_ops = {
@@ -712,10 +719,9 @@ static void exynos_tmu_work(struct work_struct *work)
 	struct exynos_tmu_data *data = container_of(work,
 			struct exynos_tmu_data, irq_work);
 
+	exynos_report_trigger();
 	mutex_lock(&data->lock);
 	clk_enable(data->clk);
-
-
 	if (data->soc == SOC_ARCH_EXYNOS)
 		writel(EXYNOS_TMU_CLEAR_RISE_INT |
 				EXYNOS_TMU_CLEAR_FALL_INT,
@@ -723,10 +729,9 @@ static void exynos_tmu_work(struct work_struct *work)
 	else
 		writel(EXYNOS4210_TMU_INTCLEAR_VAL,
 				data->base + EXYNOS_TMU_REG_INTCLEAR);
-
 	clk_disable(data->clk);
 	mutex_unlock(&data->lock);
-	exynos_report_trigger();
+
 	enable_irq(data->irq);
 }
 
-- 
1.7.5.4

Re: [PATCH 2/5 RESEND] thermal: exynos: Miscellaneous fixes to support falling threshold interrupt

[Joe Perches]

On Sun, 2013-01-06 at 15:50 -0800, Amit Daniel Kachhap wrote:
> Below fixes are done to support falling threshold interrupt,
> * Falling interrupt status macro corrected according to exynos5 data sheet.
> * The get trend function modified to calculate trip temperature correctly.
> * The clearing of interrupt status in the isr is now done after handling
>   the event that caused the interrupt.
[]
> diff --git a/drivers/thermal/exynos_thermal.c b/drivers/thermal/exynos_thermal.c
[]
> @@ -373,12 +373,19 @@ static int exynos_get_temp(struct thermal_zone_device *thermal,
>  static int exynos_get_trend(struct thermal_zone_device *thermal,
>  			int trip, enum thermal_trend *trend)
>  {
> -	if (thermal->temperature >= trip)
> +	int ret = 0;

Unnecessary initialization

> +	unsigned long trip_temp;
> +
> +	ret = exynos_get_trip_temp(thermal, trip, &trip_temp);
> +	if (ret < 0)
> +		return ret;
> +
> +	if (thermal->temperature >= trip_temp)
>  		*trend = THERMAL_TREND_RAISING;
>  	else
>  		*trend = THERMAL_TREND_DROPPING;

THERMAL_TREND_STABLE ?

>  
> -	return 0;
> +	return ret;

return 0 is clearer.

Re: [PATCH 2/5 RESEND] thermal: exynos: Miscellaneous fixes to support falling threshold interrupt

[amit kachhap]

Hi Joe,

Thanks for the review. Will re-post with your suggestion,

On Sun, Jan 6, 2013 at 3:55 PM, Joe Perches <joe@perches.com> wrote:
> On Sun, 2013-01-06 at 15:50 -0800, Amit Daniel Kachhap wrote:
>> Below fixes are done to support falling threshold interrupt,
>> * Falling interrupt status macro corrected according to exynos5 data sheet.
>> * The get trend function modified to calculate trip temperature correctly.
>> * The clearing of interrupt status in the isr is now done after handling
>>   the event that caused the interrupt.
> []
>> diff --git a/drivers/thermal/exynos_thermal.c b/drivers/thermal/exynos_thermal.c
> []
>> @@ -373,12 +373,19 @@ static int exynos_get_temp(struct thermal_zone_device *thermal,
>>  static int exynos_get_trend(struct thermal_zone_device *thermal,
>>                       int trip, enum thermal_trend *trend)
>>  {
>> -     if (thermal->temperature >= trip)
>> +     int ret = 0;
Yes agreed. Will modify it.
>
> Unnecessary initialization
>
>> +     unsigned long trip_temp;
>> +
>> +     ret = exynos_get_trip_temp(thermal, trip, &trip_temp);
>> +     if (ret < 0)
>> +             return ret;
>> +
>> +     if (thermal->temperature >= trip_temp)
>>               *trend = THERMAL_TREND_RAISING;
>>       else
>>               *trend = THERMAL_TREND_DROPPING;
>
> THERMAL_TREND_STABLE ?
Only 2 trend is sufficient. It is stable for some time as the falling
threshold interrupt is some units below the trip temp.
>
>>
>> -     return 0;
>> +     return ret;
Ok agreed
>
> return 0 is clearer.
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH V3 2/3] thermal: exynos: Miscellaneous fixes to support falling threshold interrupt

[Amit Daniel Kachhap]

Below fixes are done to support falling threshold interrupt,
* Falling interrupt status macro corrected according to exynos5 data sheet.
* The get trend function modified to calculate trip temperature correctly.
* The clearing of interrupt status in the isr is now done after handling
  the event that caused the interrupt.

Signed-off-by: Amit Daniel Kachhap <amit.daniel@samsung.com>
---
Changes since V2,
Incorporated Joe Perches <joe@perches.com> review comments about coding
guidelines.

This path is based on thermal maintainer next tree.

 drivers/thermal/exynos_thermal.c |   17 +++++++++++------
 1 files changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/thermal/exynos_thermal.c b/drivers/thermal/exynos_thermal.c
index dcd13f7..f4dd68a 100644
--- a/drivers/thermal/exynos_thermal.c
+++ b/drivers/thermal/exynos_thermal.c
@@ -82,7 +82,7 @@
 
 #define EXYNOS_TRIMINFO_RELOAD		0x1
 #define EXYNOS_TMU_CLEAR_RISE_INT	0x111
-#define EXYNOS_TMU_CLEAR_FALL_INT	(0x111 << 16)
+#define EXYNOS_TMU_CLEAR_FALL_INT	(0x111 << 12)
 #define EXYNOS_MUX_ADDR_VALUE		6
 #define EXYNOS_MUX_ADDR_SHIFT		20
 #define EXYNOS_TMU_TRIP_MODE_SHIFT	13
@@ -373,7 +373,14 @@ static int exynos_get_temp(struct thermal_zone_device *thermal,
 static int exynos_get_trend(struct thermal_zone_device *thermal,
 			int trip, enum thermal_trend *trend)
 {
-	if (thermal->temperature >= trip)
+	int ret;
+	unsigned long trip_temp;
+
+	ret = exynos_get_trip_temp(thermal, trip, &trip_temp);
+	if (ret < 0)
+		return ret;
+
+	if (thermal->temperature >= trip_temp)
 		*trend = THERMAL_TREND_RAISING;
 	else
 		*trend = THERMAL_TREND_DROPPING;
@@ -712,10 +719,9 @@ static void exynos_tmu_work(struct work_struct *work)
 	struct exynos_tmu_data *data = container_of(work,
 			struct exynos_tmu_data, irq_work);
 
+	exynos_report_trigger();
 	mutex_lock(&data->lock);
 	clk_enable(data->clk);
-
-
 	if (data->soc == SOC_ARCH_EXYNOS)
 		writel(EXYNOS_TMU_CLEAR_RISE_INT |
 				EXYNOS_TMU_CLEAR_FALL_INT,
@@ -723,10 +729,9 @@ static void exynos_tmu_work(struct work_struct *work)
 	else
 		writel(EXYNOS4210_TMU_INTCLEAR_VAL,
 				data->base + EXYNOS_TMU_REG_INTCLEAR);
-
 	clk_disable(data->clk);
 	mutex_unlock(&data->lock);
-	exynos_report_trigger();
+
 	enable_irq(data->irq);
 }
 
-- 
1.7.5.4

Re: [PATCH V3 2/3] thermal: exynos: Miscellaneous fixes to support falling threshold interrupt

[Shubhrajyoti Datta]

On 1/17/13, Amit Daniel Kachhap <amit.daniel@samsung.com> wrote:
> Below fixes are done to support falling threshold interrupt,
> * Falling interrupt status macro corrected according to exynos5 data sheet.
> * The get trend function modified to calculate trip temperature correctly.
> * The clearing of interrupt status in the isr is now done after handling
>   the event that caused the interrupt.
One fix per patch would have been better.

However not particular.

Also are they applicable to older kernels?

>
> Signed-off-by: Amit Daniel Kachhap <amit.daniel@samsung.com>
> ---