On Wed, Jan 25, 2012 at 04:34:27AM -0500, Jean Delvare wrote: > On Mon, 23 Jan 2012 18:48:44 -0800, Guenter Roeck wrote: > > When updating vrm, the value range was not limited. This could result in more or > > less random vrm values if the value provided by the user was larger than 255. > > Fix by limiting the range to 0..255 using the SENSORS_LIMIT macro. > > > > Cc: Jean Delvare <khali@xxxxxxxxxxxx> > > Signed-off-by: Guenter Roeck <guenter.roeck@xxxxxxxxxxxx> > > --- > > drivers/hwmon/lm87.c | 2 +- > > 1 files changed, 1 insertions(+), 1 deletions(-) > > > > diff --git a/drivers/hwmon/lm87.c b/drivers/hwmon/lm87.c > > index 8e66f99..2aa5e1a 100644 > > --- a/drivers/hwmon/lm87.c > > +++ b/drivers/hwmon/lm87.c > > @@ -604,7 +604,7 @@ static ssize_t set_vrm(struct device *dev, struct device_attribute *attr, > > err = kstrtoul(buf, 10, &val); > > if (err) > > return err; > > - data->vrm = val; > > + data->vrm = SENSORS_LIMIT(val, 0, 255); > > return count; > > } > > static DEVICE_ATTR(vrm, S_IRUGO | S_IWUSR, show_vrm, set_vrm); > > Same concerns here as in it87, of course. I suggest you withdraw these > patches for the time being. > Agreed, makes sense. I'll move those to my "staging" branch for now. I have a RFC set of patches in the works for hwmon-vid, trying to start addressing the vrm issue. We can go from there once we decided how to handle it. Thanks, Guenter _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
