[PATCH 2/2] thermal: db8500_cpufreq_cooling: remove usage of IS_ERR_OR_NULL()

[PATCH 2/2] thermal: db8500_cpufreq_cooling: remove usage of IS_ERR_OR_NULL()

[Eduardo Valentin]

This patch changes the driver to avoid the usage of IS_ERR_OR_NULL()
macro. This macro can lead to dangerous results, like returning
success (0) during a failure scenario (NULL pointer handling).

The case present in this driver can be translated to a simple
check for IS_ERR(), as the cpufreq_cooling_register() returns
either a valid pointer or an ERR_PTR().

Cc: Zhang Rui <rui.zhang@intel.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Grant Likely <grant.likely@secretlab.ca>
Cc: Rob Herring <rob.herring@calxeda.com>
Cc: Hongbo Zhang <hongbo.zhang@stericsson.com>
Cc: linux-pm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: devicetree-discuss@lists.ozlabs.org
Signed-off-by: Eduardo Valentin <eduardo.valentin@ti.com>
---
 drivers/thermal/db8500_cpufreq_cooling.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/thermal/db8500_cpufreq_cooling.c b/drivers/thermal/db8500_cpufreq_cooling.c
index 21419851..786d192 100644
--- a/drivers/thermal/db8500_cpufreq_cooling.c
+++ b/drivers/thermal/db8500_cpufreq_cooling.c
@@ -37,7 +37,7 @@ static int db8500_cpufreq_cooling_probe(struct platform_device *pdev)
 	cpumask_set_cpu(0, &mask_val);
 	cdev = cpufreq_cooling_register(&mask_val);
 
-	if (IS_ERR_OR_NULL(cdev)) {
+	if (IS_ERR(cdev)) {
 		dev_err(&pdev->dev, "Failed to register cooling device\n");
 		return PTR_ERR(cdev);
 	}
-- 
1.8.2.1.342.gfa7285d

Re: [PATCH 2/2] thermal: db8500_cpufreq_cooling: remove usage of IS_ERR_OR_NULL()

[Russell King]

On Thu, Apr 25, 2013 at 10:13:34AM -0400, Eduardo Valentin wrote:
> diff --git a/drivers/thermal/db8500_cpufreq_cooling.c b/drivers/thermal/db8500_cpufreq_cooling.c
> index 21419851..786d192 100644
> --- a/drivers/thermal/db8500_cpufreq_cooling.c
> +++ b/drivers/thermal/db8500_cpufreq_cooling.c
> @@ -37,7 +37,7 @@ static int db8500_cpufreq_cooling_probe(struct platform_device *pdev)
>  	cpumask_set_cpu(0, &mask_val);
>  	cdev = cpufreq_cooling_register(&mask_val);
>  
> -	if (IS_ERR_OR_NULL(cdev)) {
> +	if (IS_ERR(cdev)) {
>  		dev_err(&pdev->dev, "Failed to register cooling device\n");
>  		return PTR_ERR(cdev);

Correct.  cpufreq_cooling_register() returns either an error-pointer or
a valid pointer.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:

Re: [PATCH 2/2] thermal: db8500_cpufreq_cooling: remove usage of IS_ERR_OR_NULL()

[Eduardo Valentin]

On 25-04-2013 13:46, Russell King wrote:
> On Thu, Apr 25, 2013 at 10:13:34AM -0400, Eduardo Valentin wrote:
>> diff --git a/drivers/thermal/db8500_cpufreq_cooling.c b/drivers/thermal/db8500_cpufreq_cooling.c
>> index 21419851..786d192 100644
>> --- a/drivers/thermal/db8500_cpufreq_cooling.c
>> +++ b/drivers/thermal/db8500_cpufreq_cooling.c
>> @@ -37,7 +37,7 @@ static int db8500_cpufreq_cooling_probe(struct platform_device *pdev)
>>   	cpumask_set_cpu(0, &mask_val);
>>   	cdev = cpufreq_cooling_register(&mask_val);
>>
>> -	if (IS_ERR_OR_NULL(cdev)) {
>> +	if (IS_ERR(cdev)) {
>>   		dev_err(&pdev->dev, "Failed to register cooling device\n");
>>   		return PTR_ERR(cdev);
>
> Correct.  cpufreq_cooling_register() returns either an error-pointer or
> a valid pointer.
>

Thanks for your review!

Re: [PATCH 2/2] thermal: db8500_cpufreq_cooling: remove usage of IS_ERR_OR_NULL()

[Fabio Baltieri]

On Thu, Apr 25, 2013 at 06:46:35PM +0100, Russell King wrote:
> On Thu, Apr 25, 2013 at 10:13:34AM -0400, Eduardo Valentin wrote:
> > diff --git a/drivers/thermal/db8500_cpufreq_cooling.c b/drivers/thermal/db8500_cpufreq_cooling.c
> > index 21419851..786d192 100644
> > --- a/drivers/thermal/db8500_cpufreq_cooling.c
> > +++ b/drivers/thermal/db8500_cpufreq_cooling.c
> > @@ -37,7 +37,7 @@ static int db8500_cpufreq_cooling_probe(struct platform_device *pdev)
> >  	cpumask_set_cpu(0, &mask_val);
> >  	cdev = cpufreq_cooling_register(&mask_val);
> >  
> > -	if (IS_ERR_OR_NULL(cdev)) {
> > +	if (IS_ERR(cdev)) {
> >  		dev_err(&pdev->dev, "Failed to register cooling device\n");
> >  		return PTR_ERR(cdev);
> 
> Correct.  cpufreq_cooling_register() returns either an error-pointer or
> a valid pointer.

Acked-by: Fabio Baltieri <fabio.baltieri@linaro.org>

...but more of these are going to appear forever, was the proposed
removal of IS_ERR_OR_NULL rejected?  Can't find any new message about
that on the list.

Fabio

-- 
Fabio Baltieri

Re: [PATCH 2/2] thermal: db8500_cpufreq_cooling: remove usage of IS_ERR_OR_NULL()

[Russell King]

On Fri, Apr 26, 2013 at 12:19:08PM +0200, Fabio Baltieri wrote:
> On Thu, Apr 25, 2013 at 06:46:35PM +0100, Russell King wrote:
> > On Thu, Apr 25, 2013 at 10:13:34AM -0400, Eduardo Valentin wrote:
> > > diff --git a/drivers/thermal/db8500_cpufreq_cooling.c b/drivers/thermal/db8500_cpufreq_cooling.c
> > > index 21419851..786d192 100644
> > > --- a/drivers/thermal/db8500_cpufreq_cooling.c
> > > +++ b/drivers/thermal/db8500_cpufreq_cooling.c
> > > @@ -37,7 +37,7 @@ static int db8500_cpufreq_cooling_probe(struct platform_device *pdev)
> > >  	cpumask_set_cpu(0, &mask_val);
> > >  	cdev = cpufreq_cooling_register(&mask_val);
> > >  
> > > -	if (IS_ERR_OR_NULL(cdev)) {
> > > +	if (IS_ERR(cdev)) {
> > >  		dev_err(&pdev->dev, "Failed to register cooling device\n");
> > >  		return PTR_ERR(cdev);
> > 
> > Correct.  cpufreq_cooling_register() returns either an error-pointer or
> > a valid pointer.
> 
> Acked-by: Fabio Baltieri <fabio.baltieri@linaro.org>
> 
> ...but more of these are going to appear forever, was the proposed
> removal of IS_ERR_OR_NULL rejected?  Can't find any new message about
> that on the list.

We first need to get rid of the existing users of it - I've got rid of
most of those in arch/arm - but it seems that I never pushed that in the
last merge window.

Really it needs the cooperation of everyone to make it happen across the
tree with everyone removing IS_ERR_OR_NULL() in their subsystem.  There
are currently 366 instances of this macro being used in the entire tree
which is far too many to even mark it as deprecated.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of: