[lm-sensors] [PATCH] hwmon: (adt7473) Initialize max_duty_at_overheat before use
khali at linux-fr.org
Mon Apr 28 13:17:58 CEST 2008
Thanks for the review.
On Mon, 28 Apr 2008 11:31:22 +0300, Paulius Zaleckas wrote:
> Jean Delvare wrote:
> > --- linux-2.6.26-rc0.orig/drivers/hwmon/adt7473.c 2008-04-26 15:36:41.000000000 +0200
> > +++ linux-2.6.26-rc0/drivers/hwmon/adt7473.c 2008-04-26 16:15:05.000000000 +0200
> > @@ -298,6 +298,9 @@ no_sensor_update:
> > ADT7473_REG_PWM_BHVR(i));
> > }
> > + i = i2c_smbus_read_byte_data(client, ADT7473_REG_CFG4);
> > + data->max_duty_at_overheat = !!(i & ADT7473_CFG4_MAX_DUTY_AT_OVT);
> Maybe it would be more understandable to use:
> data->max_duty_at_overheat = (i & ADT7473_CFG4_MAX_DUTY_AT_OVT) ? 1 : 0;
The !! construct seems to be widely used in the kernel, I found 800
occurrences, so I guess it's OK to use it here. But if Darrick (as the
driver author) or Mark (as the subsystem maintainer) prefer your
variant, I can update the patch.
> Although I don't know if gcc optimizes both to the same instructions.
It seems so, yes (gcc 4.1.2).
> > +
> > data->limits_last_updated = local_jiffies;
> > data->limits_valid = 1;
More information about the lm-sensors