public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Guenter Roeck <guenter.roeck@ericsson.com>
To: Clemens Ladisch <clemens@ladisch.de>
Cc: Guenter Roeck <linux@roeck-us.net>,
	Jean Delvare <khali@linux-fr.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"lm-sensors@lm-sensors.org" <lm-sensors@lm-sensors.org>
Subject: Re: [lm-sensors] [PATCH 3/3] hwmon: (jc42) do not allow writing to locked registers
Date: Wed, 16 Feb 2011 06:50:49 -0800	[thread overview]
Message-ID: <20110216145049.GC13872@ericsson.com> (raw)
In-Reply-To: <4D5BCAEE.6030502@ladisch.de>

Hi Clemens,

On Wed, Feb 16, 2011 at 08:02:38AM -0500, Clemens Ladisch wrote:
> On systems where the temperature sensor is actually used, the BIOS is
> likely to have locked the alarm registers.  In that case, all writes
> through the corresponding sysfs files would be silently ignored.
> 
> To prevent this, detect the locks and make the affected sysfs files
> read-only.
> 
> Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
> ---
>  Documentation/hwmon/jc42 |   12 ++++++++----
>  drivers/hwmon/jc42.c     |   33 +++++++++++++++++++++++++++++----
>  2 files changed, 37 insertions(+), 8 deletions(-)
> 
> --- a/Documentation/hwmon/jc42
> +++ b/Documentation/hwmon/jc42
> @@ -86,15 +86,19 @@ limits. The chip supports only a single
>  which applies to all limits. This register can be written by writing into
>  temp1_crit_hyst. Other hysteresis attributes are read-only.
>  
> +If the BIOS has configured the sensor for automatic temperature management, it
> +is likely that it has locked the registers, i.e., that the temperature limits
> +cannot be changed.
> +
>  Sysfs entries
>  -------------
>  
>  temp1_input		Temperature (RO)
> -temp1_min		Minimum temperature (RW)
> -temp1_max		Maximum temperature (RW)
> -temp1_crit		Critical high temperature (RW)
> +temp1_min		Minimum temperature (RO or RW)
> +temp1_max		Maximum temperature (RO or RW)
> +temp1_crit		Critical high temperature (RO or RW)
>  
> -temp1_crit_hyst		Critical hysteresis temperature (RW)
> +temp1_crit_hyst		Critical hysteresis temperature (RO or RW)
>  temp1_max_hyst		Maximum hysteresis temperature (RO)
>  
>  temp1_min_alarm		Temperature low alarm
> --- a/drivers/hwmon/jc42.c
> +++ b/drivers/hwmon/jc42.c
> @@ -53,6 +53,8 @@ static const unsigned short normal_i2c[]
>  
>  /* Configuration register defines */
>  #define JC42_CFG_CRIT_ONLY	(1 << 2)
> +#define JC42_CFG_TCRIT_LOCK	(1 << 6)
> +#define JC42_CFG_EVENT_LOCK	(1 << 7)
>  #define JC42_CFG_SHUTDOWN	(1 << 8)
>  #define JC42_CFG_HYST_SHIFT	9
>  #define JC42_CFG_HYST_MASK	0x03
> @@ -380,14 +382,14 @@ static ssize_t show_alarm(struct device
>  
>  static DEVICE_ATTR(temp1_input, S_IRUGO,
>  		   show_temp_input, NULL);
> -static DEVICE_ATTR(temp1_crit, S_IWUSR | S_IRUGO,
> +static DEVICE_ATTR(temp1_crit, S_IRUGO,
>  		   show_temp_crit, set_temp_crit);
> -static DEVICE_ATTR(temp1_min, S_IWUSR | S_IRUGO,
> +static DEVICE_ATTR(temp1_min, S_IRUGO,
>  		   show_temp_min, set_temp_min);
> -static DEVICE_ATTR(temp1_max, S_IWUSR | S_IRUGO,
> +static DEVICE_ATTR(temp1_max, S_IRUGO,
>  		   show_temp_max, set_temp_max);
>  
> -static DEVICE_ATTR(temp1_crit_hyst, S_IWUSR | S_IRUGO,
> +static DEVICE_ATTR(temp1_crit_hyst, S_IRUGO,
>  		   show_temp_crit_hyst, set_temp_crit_hyst);
>  static DEVICE_ATTR(temp1_max_hyst, S_IRUGO,
>  		   show_temp_max_hyst, NULL);
> @@ -412,8 +414,31 @@ static struct attribute *jc42_attributes
>  	NULL
>  };
>  
> +static mode_t jc42_attribute_mode(struct kobject *kobj,
> +				  struct attribute *attr, int index)
> +{
> +	struct device *dev = container_of(kobj, struct device, kobj);
> +	struct i2c_client *client = to_i2c_client(dev);
> +	struct jc42_data *data = i2c_get_clientdata(client);
> +	unsigned int config = data->config;
> +	bool readonly;
> +
> +	if (attr == &dev_attr_temp1_crit.attr)
> +		readonly = config & JC42_CFG_TCRIT_LOCK;

You are assigning a non-bool to a bool. I can see that recent C compilers
do the right thing, but I am not sure if that is always the case.
So I would prefer
		readonly = !!(config & JC42_CFG_TCRIT_LOCK));

Same for the assignments below. I can make that change if you are ok with it.

> +	else if (attr == &dev_attr_temp1_min.attr ||
> +		 attr == &dev_attr_temp1_max.attr)
> +		readonly = config & JC42_CFG_EVENT_LOCK;
> +	else if (attr == &dev_attr_temp1_crit_hyst.attr)
> +		readonly = config & (JC42_CFG_EVENT_LOCK | JC42_CFG_TCRIT_LOCK);
> +	else
> +		readonly = true;
> +
> +	return S_IRUGO | (readonly ? 0 : S_IWUSR);
> +}
> +
>  static const struct attribute_group jc42_group = {
>  	.attrs = jc42_attributes,
> +	.is_visible = jc42_attribute_mode,
>  };
>  
>  /* Return 0 if detection is successful, -ENODEV otherwise */
> 
> _______________________________________________
> lm-sensors mailing list
> lm-sensors@lm-sensors.org
> http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

  reply	other threads:[~2011-02-16 14:51 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-02-16 13:00 [PATCH 0/3] hwmon: some jc42 changes Clemens Ladisch
2011-02-16 13:01 ` [PATCH 1/3] hwmon: (jc42) fix type mismatch Clemens Ladisch
2011-02-16 15:09   ` [lm-sensors] " Guenter Roeck
2011-02-16 13:02 ` [PATCH 2/3] hwmon: (jc42) more helpful documentation Clemens Ladisch
2011-02-16 15:09   ` [lm-sensors] " Guenter Roeck
2011-02-16 13:02 ` [PATCH 3/3] hwmon: (jc42) do not allow writing to locked registers Clemens Ladisch
2011-02-16 14:50   ` Guenter Roeck [this message]
2011-02-16 15:11     ` [lm-sensors] " Clemens Ladisch
2011-02-16 15:20       ` Guenter Roeck
2011-02-16 15:38         ` Clemens Ladisch
2011-02-16 16:15           ` Guenter Roeck

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110216145049.GC13872@ericsson.com \
    --to=guenter.roeck@ericsson.com \
    --cc=clemens@ladisch.de \
    --cc=khali@linux-fr.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@roeck-us.net \
    --cc=lm-sensors@lm-sensors.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox