On Wed, Jun 20, 2012 at 04:52:01PM +0200, Jean Delvare wrote: > On Tue, 19 Jun 2012 08:46:42 -0700, Guenter Roeck wrote: > > An unsigned value is never smaller than 0. Remove the check for it. > > > > Signed-off-by: Guenter Roeck <linux@xxxxxxxxxxxx> > > --- > > drivers/hwmon/acpi_power_meter.c | 2 -- > > 1 file changed, 2 deletions(-) > > > > diff --git a/drivers/hwmon/acpi_power_meter.c b/drivers/hwmon/acpi_power_meter.c > > index 34ad5a2..e2b0ae8 100644 > > --- a/drivers/hwmon/acpi_power_meter.c > > +++ b/drivers/hwmon/acpi_power_meter.c > > @@ -308,8 +308,6 @@ static ssize_t set_trip(struct device *dev, struct device_attribute *devattr, > > return res; > > > > temp /= 1000; > > - if (temp < 0) > > - return -EINVAL; > > > > mutex_lock(&resource->lock); > > resource->trip[attr->index - 7] = temp; > > Acked-by: Jean Delvare <khali@xxxxxxxxxxxx> > > That being said, DIV_ROUND_CLOSEST would be better to use than /=, and > some clamping on the value would be good to have too, as you can't > always fit an unsigned long into an int (there may even be other > constraints, I don't know.) I thought about doing that as separate patch, but since the above doesn't really fix a bug anyway, it does not hurt to put it together into one. I'll do that and resubmit. Guenter _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
