[PATCH 1/1] thermal: fix cpu_cooling max_level behavior

[PATCH 1/1] thermal: fix cpu_cooling max_level behavior

[Eduardo Valentin]

As per Documentation/thermal/sysfs-api.txt, max_level
is an index, not a counter. Thus, in case a CPU has
3 valid frequencies, max_level is expected to be 2, for instance.

The current code makes max_level == number of valid frequencies,
which is bogus. This patch fix the cpu_cooling device by
ranging max_level properly.

Reported-by: Carlos Hernandez <ceh@ti.com>
Signed-off-by: Eduardo Valentin <eduardo.valentin@ti.com>
---
 drivers/thermal/cpu_cooling.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
index d179028..d0f8f8b5 100644
--- a/drivers/thermal/cpu_cooling.c
+++ b/drivers/thermal/cpu_cooling.c
@@ -173,6 +173,8 @@ static int get_property(unsigned int cpu, unsigned long input,
 		freq = table[i].frequency;
 		max_level++;
 	}
+	/* max_level is an index, not a counter */
+	max_level--;
 
 	/* get max level */
 	if (property == GET_MAXL) {
@@ -181,7 +183,7 @@ static int get_property(unsigned int cpu, unsigned long input,
 	}
 
 	if (property == GET_FREQ)
-		level = descend ? input : (max_level - input - 1);
+		level = descend ? input : (max_level - input);
 
 	for (i = 0, j = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) {
 		/* ignore invalid entry */
@@ -197,7 +199,7 @@ static int get_property(unsigned int cpu, unsigned long input,
 
 		if (property == GET_LEVEL && (unsigned int)input == freq) {
 			/* get level by frequency */
-			*output = descend ? j : (max_level - j - 1);
+			*output = descend ? j : (max_level - j);
 			return 0;
 		}
 		if (property == GET_FREQ && level == j) {
-- 
1.8.2.1.342.gfa7285d

[PATCH 1/1] thermal: fix cpu_cooling max_level behavior

[Eduardo Valentin]

As per Documentation/thermal/sysfs-api.txt, max_level
is an index, not a counter. Thus, in case a CPU has
3 valid frequencies, max_level is expected to be 2, for instance.

The current code makes max_level == number of valid frequencies,
which is bogus. This patch fix the cpu_cooling device by
ranging max_level properly.

Reported-by: Carlos Hernandez <ceh@ti.com>
Signed-off-by: Eduardo Valentin <eduardo.valentin@ti.com>
---
 drivers/thermal/cpu_cooling.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
index d179028..d0f8f8b5 100644
--- a/drivers/thermal/cpu_cooling.c
+++ b/drivers/thermal/cpu_cooling.c
@@ -173,6 +173,8 @@ static int get_property(unsigned int cpu, unsigned long input,
 		freq = table[i].frequency;
 		max_level++;
 	}
+	/* max_level is an index, not a counter */
+	max_level--;
 
 	/* get max level */
 	if (property == GET_MAXL) {
@@ -181,7 +183,7 @@ static int get_property(unsigned int cpu, unsigned long input,
 	}
 
 	if (property == GET_FREQ)
-		level = descend ? input : (max_level - input - 1);
+		level = descend ? input : (max_level - input);
 
 	for (i = 0, j = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) {
 		/* ignore invalid entry */
@@ -197,7 +199,7 @@ static int get_property(unsigned int cpu, unsigned long input,
 
 		if (property == GET_LEVEL && (unsigned int)input == freq) {
 			/* get level by frequency */
-			*output = descend ? j : (max_level - j - 1);
+			*output = descend ? j : (max_level - j);
 			return 0;
 		}
 		if (property == GET_FREQ && level == j) {
-- 
1.8.2.1.342.gfa7285d

Re: [PATCH 1/1] thermal: fix cpu_cooling max_level behavior

[Eduardo Valentin]

[-- Attachment #1: Type: text/plain, Size: 1940 bytes --]

On 13-11-2013 14:11, Eduardo Valentin wrote:
> As per Documentation/thermal/sysfs-api.txt, max_level
> is an index, not a counter. Thus, in case a CPU has
> 3 valid frequencies, max_level is expected to be 2, for instance.
> 
> The current code makes max_level == number of valid frequencies,
> which is bogus. This patch fix the cpu_cooling device by
> ranging max_level properly.
> 
> Reported-by: Carlos Hernandez <ceh@ti.com>
> Signed-off-by: Eduardo Valentin <eduardo.valentin@ti.com>

Rui,

This fix is applicable even on 3.12.

> ---
>  drivers/thermal/cpu_cooling.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
> index d179028..d0f8f8b5 100644
> --- a/drivers/thermal/cpu_cooling.c
> +++ b/drivers/thermal/cpu_cooling.c
> @@ -173,6 +173,8 @@ static int get_property(unsigned int cpu, unsigned long input,
>  		freq = table[i].frequency;
>  		max_level++;
>  	}
> +	/* max_level is an index, not a counter */
> +	max_level--;
>  
>  	/* get max level */
>  	if (property == GET_MAXL) {
> @@ -181,7 +183,7 @@ static int get_property(unsigned int cpu, unsigned long input,
>  	}
>  
>  	if (property == GET_FREQ)
> -		level = descend ? input : (max_level - input - 1);
> +		level = descend ? input : (max_level - input);
>  
>  	for (i = 0, j = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) {
>  		/* ignore invalid entry */
> @@ -197,7 +199,7 @@ static int get_property(unsigned int cpu, unsigned long input,
>  
>  		if (property == GET_LEVEL && (unsigned int)input == freq) {
>  			/* get level by frequency */
> -			*output = descend ? j : (max_level - j - 1);
> +			*output = descend ? j : (max_level - j);
>  			return 0;
>  		}
>  		if (property == GET_FREQ && level == j) {
> 


-- 
You have got to be excited about what you are doing. (L. Lamport)

Eduardo Valentin


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 295 bytes --]

Re: [PATCH 1/1] thermal: fix cpu_cooling max_level behavior

[Eduardo Valentin]

[-- Attachment #1: Type: text/plain, Size: 1940 bytes --]

On 13-11-2013 14:11, Eduardo Valentin wrote:
> As per Documentation/thermal/sysfs-api.txt, max_level
> is an index, not a counter. Thus, in case a CPU has
> 3 valid frequencies, max_level is expected to be 2, for instance.
> 
> The current code makes max_level == number of valid frequencies,
> which is bogus. This patch fix the cpu_cooling device by
> ranging max_level properly.
> 
> Reported-by: Carlos Hernandez <ceh@ti.com>
> Signed-off-by: Eduardo Valentin <eduardo.valentin@ti.com>

Rui,

This fix is applicable even on 3.12.

> ---
>  drivers/thermal/cpu_cooling.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
> index d179028..d0f8f8b5 100644
> --- a/drivers/thermal/cpu_cooling.c
> +++ b/drivers/thermal/cpu_cooling.c
> @@ -173,6 +173,8 @@ static int get_property(unsigned int cpu, unsigned long input,
>  		freq = table[i].frequency;
>  		max_level++;
>  	}
> +	/* max_level is an index, not a counter */
> +	max_level--;
>  
>  	/* get max level */
>  	if (property == GET_MAXL) {
> @@ -181,7 +183,7 @@ static int get_property(unsigned int cpu, unsigned long input,
>  	}
>  
>  	if (property == GET_FREQ)
> -		level = descend ? input : (max_level - input - 1);
> +		level = descend ? input : (max_level - input);
>  
>  	for (i = 0, j = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) {
>  		/* ignore invalid entry */
> @@ -197,7 +199,7 @@ static int get_property(unsigned int cpu, unsigned long input,
>  
>  		if (property == GET_LEVEL && (unsigned int)input == freq) {
>  			/* get level by frequency */
> -			*output = descend ? j : (max_level - j - 1);
> +			*output = descend ? j : (max_level - j);
>  			return 0;
>  		}
>  		if (property == GET_FREQ && level == j) {
> 


-- 
You have got to be excited about what you are doing. (L. Lamport)

Eduardo Valentin


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 295 bytes --]

Re: [PATCH 1/1] thermal: fix cpu_cooling max_level behavior

[Eduardo Valentin]

[-- Attachment #1: Type: text/plain, Size: 1945 bytes --]

On 13-11-2013 14:11, Eduardo Valentin wrote:
> As per Documentation/thermal/sysfs-api.txt, max_level
> is an index, not a counter. Thus, in case a CPU has
> 3 valid frequencies, max_level is expected to be 2, for instance.
> 
> The current code makes max_level == number of valid frequencies,
> which is bogus. This patch fix the cpu_cooling device by
> ranging max_level properly.
> 
> Reported-by: Carlos Hernandez <ceh@ti.com>
> Signed-off-by: Eduardo Valentin <eduardo.valentin@ti.com>

Rui,

Can you please consider pushing this fix?

> ---
>  drivers/thermal/cpu_cooling.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
> index d179028..d0f8f8b5 100644
> --- a/drivers/thermal/cpu_cooling.c
> +++ b/drivers/thermal/cpu_cooling.c
> @@ -173,6 +173,8 @@ static int get_property(unsigned int cpu, unsigned long input,
>  		freq = table[i].frequency;
>  		max_level++;
>  	}
> +	/* max_level is an index, not a counter */
> +	max_level--;
>  
>  	/* get max level */
>  	if (property == GET_MAXL) {
> @@ -181,7 +183,7 @@ static int get_property(unsigned int cpu, unsigned long input,
>  	}
>  
>  	if (property == GET_FREQ)
> -		level = descend ? input : (max_level - input - 1);
> +		level = descend ? input : (max_level - input);
>  
>  	for (i = 0, j = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) {
>  		/* ignore invalid entry */
> @@ -197,7 +199,7 @@ static int get_property(unsigned int cpu, unsigned long input,
>  
>  		if (property == GET_LEVEL && (unsigned int)input == freq) {
>  			/* get level by frequency */
> -			*output = descend ? j : (max_level - j - 1);
> +			*output = descend ? j : (max_level - j);
>  			return 0;
>  		}
>  		if (property == GET_FREQ && level == j) {
> 


-- 
You have got to be excited about what you are doing. (L. Lamport)

Eduardo Valentin


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 295 bytes --]

Re: [PATCH 1/1] thermal: fix cpu_cooling max_level behavior

[Eduardo Valentin]

[-- Attachment #1: Type: text/plain, Size: 1945 bytes --]

On 13-11-2013 14:11, Eduardo Valentin wrote:
> As per Documentation/thermal/sysfs-api.txt, max_level
> is an index, not a counter. Thus, in case a CPU has
> 3 valid frequencies, max_level is expected to be 2, for instance.
> 
> The current code makes max_level == number of valid frequencies,
> which is bogus. This patch fix the cpu_cooling device by
> ranging max_level properly.
> 
> Reported-by: Carlos Hernandez <ceh@ti.com>
> Signed-off-by: Eduardo Valentin <eduardo.valentin@ti.com>

Rui,

Can you please consider pushing this fix?

> ---
>  drivers/thermal/cpu_cooling.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
> index d179028..d0f8f8b5 100644
> --- a/drivers/thermal/cpu_cooling.c
> +++ b/drivers/thermal/cpu_cooling.c
> @@ -173,6 +173,8 @@ static int get_property(unsigned int cpu, unsigned long input,
>  		freq = table[i].frequency;
>  		max_level++;
>  	}
> +	/* max_level is an index, not a counter */
> +	max_level--;
>  
>  	/* get max level */
>  	if (property == GET_MAXL) {
> @@ -181,7 +183,7 @@ static int get_property(unsigned int cpu, unsigned long input,
>  	}
>  
>  	if (property == GET_FREQ)
> -		level = descend ? input : (max_level - input - 1);
> +		level = descend ? input : (max_level - input);
>  
>  	for (i = 0, j = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) {
>  		/* ignore invalid entry */
> @@ -197,7 +199,7 @@ static int get_property(unsigned int cpu, unsigned long input,
>  
>  		if (property == GET_LEVEL && (unsigned int)input == freq) {
>  			/* get level by frequency */
> -			*output = descend ? j : (max_level - j - 1);
> +			*output = descend ? j : (max_level - j);
>  			return 0;
>  		}
>  		if (property == GET_FREQ && level == j) {
> 


-- 
You have got to be excited about what you are doing. (L. Lamport)

Eduardo Valentin


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 295 bytes --]

Re: [PATCH 1/1] thermal: fix cpu_cooling max_level behavior

[Zhang Rui]

On Fri, 2013-12-06 at 09:52 -0400, Eduardo Valentin wrote:
> On 13-11-2013 14:11, Eduardo Valentin wrote:
> > As per Documentation/thermal/sysfs-api.txt, max_level
> > is an index, not a counter. Thus, in case a CPU has
> > 3 valid frequencies, max_level is expected to be 2, for instance.
> > 
> > The current code makes max_level == number of valid frequencies,
> > which is bogus. This patch fix the cpu_cooling device by
> > ranging max_level properly.
> > 
good catch.
> > Reported-by: Carlos Hernandez <ceh@ti.com>
> > Signed-off-by: Eduardo Valentin <eduardo.valentin@ti.com>
> 
> Rui,
> 
> Can you please consider pushing this fix?
> 
applied.

> > ---
> >  drivers/thermal/cpu_cooling.c | 6 ++++--
> >  1 file changed, 4 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
> > index d179028..d0f8f8b5 100644
> > --- a/drivers/thermal/cpu_cooling.c
> > +++ b/drivers/thermal/cpu_cooling.c
> > @@ -173,6 +173,8 @@ static int get_property(unsigned int cpu, unsigned long input,
> >  		freq = table[i].frequency;
> >  		max_level++;
> >  	}
> > +	/* max_level is an index, not a counter */
> > +	max_level--;
> >
I think we should check the max_level first, like the patch I attached
below.

thanks,
rui

>From a116776f7b6052599df0c67db29c30ea9d69d7ee Mon Sep 17 00:00:00 2001
From: Zhang Rui <rui.zhang@intel.com>
Date: Thu, 2 Jan 2014 11:57:48 +0800
Subject: [PATCH] Thermal cpu cooling: return error if no valid cpu frequency
 entry

Signed-off-by: Zhang Rui <rui.zhang@intel.com>
---
 drivers/thermal/cpu_cooling.c |    5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
index cc556a8..bb486b4 100644
--- a/drivers/thermal/cpu_cooling.c
+++ b/drivers/thermal/cpu_cooling.c
@@ -173,6 +173,11 @@ static int get_property(unsigned int cpu, unsigned long input,
 		freq = table[i].frequency;
 		max_level++;
 	}
+
+	/* No valid cpu frequency entry */
+	if (max_level == 0)
+		return -EINVAL;
+
 	/* max_level is an index, not a counter */
 	max_level--;
 
-- 
1.7.9.5

Re: [PATCH 1/1] thermal: fix cpu_cooling max_level behavior

[Eduardo Valentin]

[-- Attachment #1: Type: text/plain, Size: 2460 bytes --]

On 02-01-2014 00:02, Zhang Rui wrote:
> On Fri, 2013-12-06 at 09:52 -0400, Eduardo Valentin wrote:
>> On 13-11-2013 14:11, Eduardo Valentin wrote:
>>> As per Documentation/thermal/sysfs-api.txt, max_level
>>> is an index, not a counter. Thus, in case a CPU has
>>> 3 valid frequencies, max_level is expected to be 2, for instance.
>>>
>>> The current code makes max_level == number of valid frequencies,
>>> which is bogus. This patch fix the cpu_cooling device by
>>> ranging max_level properly.
>>>
> good catch.
>>> Reported-by: Carlos Hernandez <ceh@ti.com>
>>> Signed-off-by: Eduardo Valentin <eduardo.valentin@ti.com>
>>
>> Rui,
>>
>> Can you please consider pushing this fix?
>>
> applied.
> 
>>> ---
>>>  drivers/thermal/cpu_cooling.c | 6 ++++--
>>>  1 file changed, 4 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
>>> index d179028..d0f8f8b5 100644
>>> --- a/drivers/thermal/cpu_cooling.c
>>> +++ b/drivers/thermal/cpu_cooling.c
>>> @@ -173,6 +173,8 @@ static int get_property(unsigned int cpu, unsigned long input,
>>>  		freq = table[i].frequency;
>>>  		max_level++;
>>>  	}
>>> +	/* max_level is an index, not a counter */
>>> +	max_level--;
>>>
> I think we should check the max_level first, like the patch I attached
> below.
> 
> thanks,
> rui
> 
>>From a116776f7b6052599df0c67db29c30ea9d69d7ee Mon Sep 17 00:00:00 2001
> From: Zhang Rui <rui.zhang@intel.com>
> Date: Thu, 2 Jan 2014 11:57:48 +0800
> Subject: [PATCH] Thermal cpu cooling: return error if no valid cpu frequency
>  entry
> 
> Signed-off-by: Zhang Rui <rui.zhang@intel.com>
> ---
>  drivers/thermal/cpu_cooling.c |    5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
> index cc556a8..bb486b4 100644
> --- a/drivers/thermal/cpu_cooling.c
> +++ b/drivers/thermal/cpu_cooling.c
> @@ -173,6 +173,11 @@ static int get_property(unsigned int cpu, unsigned long input,
>  		freq = table[i].frequency;
>  		max_level++;
>  	}
> +
> +	/* No valid cpu frequency entry */
> +	if (max_level == 0)
> +		return -EINVAL;
> +

Agreed. For the patch above:

Acked-by: Eduardo Valentin <eduardo.valentin@ti.com>

>  	/* max_level is an index, not a counter */
>  	max_level--;
>  
> 


-- 
You have got to be excited about what you are doing. (L. Lamport)

Eduardo Valentin


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 295 bytes --]

Re: [PATCH 1/1] thermal: fix cpu_cooling max_level behavior

[Eduardo Valentin]

[-- Attachment #1: Type: text/plain, Size: 2460 bytes --]

On 02-01-2014 00:02, Zhang Rui wrote:
> On Fri, 2013-12-06 at 09:52 -0400, Eduardo Valentin wrote:
>> On 13-11-2013 14:11, Eduardo Valentin wrote:
>>> As per Documentation/thermal/sysfs-api.txt, max_level
>>> is an index, not a counter. Thus, in case a CPU has
>>> 3 valid frequencies, max_level is expected to be 2, for instance.
>>>
>>> The current code makes max_level == number of valid frequencies,
>>> which is bogus. This patch fix the cpu_cooling device by
>>> ranging max_level properly.
>>>
> good catch.
>>> Reported-by: Carlos Hernandez <ceh@ti.com>
>>> Signed-off-by: Eduardo Valentin <eduardo.valentin@ti.com>
>>
>> Rui,
>>
>> Can you please consider pushing this fix?
>>
> applied.
> 
>>> ---
>>>  drivers/thermal/cpu_cooling.c | 6 ++++--
>>>  1 file changed, 4 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
>>> index d179028..d0f8f8b5 100644
>>> --- a/drivers/thermal/cpu_cooling.c
>>> +++ b/drivers/thermal/cpu_cooling.c
>>> @@ -173,6 +173,8 @@ static int get_property(unsigned int cpu, unsigned long input,
>>>  		freq = table[i].frequency;
>>>  		max_level++;
>>>  	}
>>> +	/* max_level is an index, not a counter */
>>> +	max_level--;
>>>
> I think we should check the max_level first, like the patch I attached
> below.
> 
> thanks,
> rui
> 
>>From a116776f7b6052599df0c67db29c30ea9d69d7ee Mon Sep 17 00:00:00 2001
> From: Zhang Rui <rui.zhang@intel.com>
> Date: Thu, 2 Jan 2014 11:57:48 +0800
> Subject: [PATCH] Thermal cpu cooling: return error if no valid cpu frequency
>  entry
> 
> Signed-off-by: Zhang Rui <rui.zhang@intel.com>
> ---
>  drivers/thermal/cpu_cooling.c |    5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
> index cc556a8..bb486b4 100644
> --- a/drivers/thermal/cpu_cooling.c
> +++ b/drivers/thermal/cpu_cooling.c
> @@ -173,6 +173,11 @@ static int get_property(unsigned int cpu, unsigned long input,
>  		freq = table[i].frequency;
>  		max_level++;
>  	}
> +
> +	/* No valid cpu frequency entry */
> +	if (max_level == 0)
> +		return -EINVAL;
> +

Agreed. For the patch above:

Acked-by: Eduardo Valentin <eduardo.valentin@ti.com>

>  	/* max_level is an index, not a counter */
>  	max_level--;
>  
> 


-- 
You have got to be excited about what you are doing. (L. Lamport)

Eduardo Valentin


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 295 bytes --]