From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eduardo Valentin <edubezval@gmail.com>
Date: Wed, 04 Nov 2015 05:57:00 +0000
Subject: Re: [patch] thermal: underflow in trip_point_temp_store()
Message-Id: <20151104055659.GB8850@localhost.localdomain>
List-Id: <kernel-janitors.vger.kernel.org>
References: <20151103221434.GB19280@mwanda>
In-Reply-To: <20151103221434.GB19280@mwanda>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: Dan Carpenter <dan.carpenter@oracle.com>
Cc: Zhang Rui <rui.zhang@intel.com>, linux-pm@vger.kernel.org, kernel-janitors@vger.kernel.org

Hello Dan,

On Wed, Nov 04, 2015 at 01:14:34AM +0300, Dan Carpenter wrote:
> This is to address a static checker warning about an underflow in
> imx_set_trip_temp().  The checker is complaining that we have a user
> supplied value for "temp" from kstrtoul() where we treat it as signed,
> we cap the upper but we accept negative values.
> 
> This looks unintentional since the caller is using unsigned longs to
> represent the temperature.  Let's change it to int and reject negatives
> in the caller.
> 
> Also I changed it to reject negative "trip" values as well.
> 
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
> ---
> Someday we will use super cooled CPUs and we will need to rethink this
> code.  :)
> 

I wish cpus would be the only type of devices covered here :-)


> diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c
> index d9e525c..151a630 100644
> --- a/drivers/thermal/thermal_core.c
> +++ b/drivers/thermal/thermal_core.c
> @@ -664,7 +664,7 @@ trip_point_temp_store(struct device *dev, struct device_attribute *attr,
>  {
>  	struct thermal_zone_device *tz = to_thermal_zone(dev);
>  	int trip, ret;
> -	unsigned long temperature;
> +	int temperature;
>  
>  	if (!tz->ops->set_trip_temp)
>  		return -EPERM;
> @@ -672,7 +672,9 @@ trip_point_temp_store(struct device *dev, struct device_attribute *attr,
>  	if (!sscanf(attr->attr.name, "trip_point_%d_temp", &trip))
>  		return -EINVAL;
>  
> -	if (kstrtoul(buf, 10, &temperature))
> +	if (kstrtoint(buf, 10, &temperature))
> +		return -EINVAL;
> +	if (trip < 0 || temperature < 0)

While this maintains the original code semantics, I would prefer we
could accept negative values here. Main reason is to maintain the range
accepted by the current type in use for temperature. Second reason is
that I heard reports of people willing to use this code to control
thermal zones that would be at temperatures below 0.

Also, while here, could you please help cleaning emul_temp_store?

Thanks for sending this patch.

BR,

Eduardo

>  		return -EINVAL;
>  
>  	ret = tz->ops->set_trip_temp(tz, trip, temperature);