Hi Guenter, On Thu, 9 Jun 2011 18:14:07 -0700, Guenter Roeck wrote: > MAX1668 and compatibles have several external temperature sensors, but only > a single FAULT status bit. If a fault occurs, the temperature reported on > affected sensors is 127 degrees C. Use this knowledge to report faults on > individual external sensors. > > Signed-off-by: Guenter Roeck <guenter.roeck@xxxxxxxxxxxx> > --- > Tested with MAX1668 with one sensor disconnected. Resulting output is as follows. > Seems to be better than just reporting extra-high temperature. Looks overall good. Acked-by: Jean Delvare <khali@xxxxxxxxxxxx> Minor comments below if you are interested. As a side note, it seems that http://git.kernel.org/?p=linux/kernel/git/groeck/staging.git;a=commit;h=af2dcfe730577d5399973a4df648190f5562b470 doesn't properly credit David as being the original author of the commit. > > root@groeck-desktop# sensors > max1668-i2c-5-18 > Adapter: i2c-devantech-iss at bus 001 device 007 > temp1: +23.0°C (low = -55.0°C, high = +127.0°C) > temp2: +25.0°C (low = -55.0°C, high = +127.0°C) > temp3: +25.0°C (low = -55.0°C, high = +127.0°C) > temp4: +25.0°C (low = -55.0°C, high = +127.0°C) > temp5: FAULT (low = -55.0°C, high = +127.0°C) ALARM(MAX) > > drivers/hwmon/max1668.c | 26 +++++++++++++++++++++++++- > 1 files changed, 25 insertions(+), 1 deletions(-) > > diff --git a/drivers/hwmon/max1668.c b/drivers/hwmon/max1668.c > index ddefb64..af85d44 100644 > --- a/drivers/hwmon/max1668.c > +++ b/drivers/hwmon/max1668.c > @@ -123,7 +123,7 @@ static struct max1668_data *max1668_update_device(struct device *dev) > goto abort; > } > data->alarms &= 0x00ff; > - data->alarms |= ((u8) (val & 0x60)) << 8; > + data->alarms |= ((u8) (val & 0x70)) << 8; The (u8) type cast doesn't seem needed. Same a few lines below for MAX1668_REG_STAT2. > > val = i2c_smbus_read_byte_data(client, MAX1668_REG_STAT2); > if (unlikely(val < 0)) { > @@ -189,6 +189,19 @@ static ssize_t show_alarm(struct device *dev, struct device_attribute *attr, > return sprintf(buf, "%u\n", (data->alarms >> index) & 0x1); > } > > +static ssize_t show_fault(struct device *dev, > + struct device_attribute *devattr, char *buf) > +{ > + int index = to_sensor_dev_attr(devattr)->index; > + struct max1668_data *data = max1668_update_device(dev); > + > + if (IS_ERR(data)) > + return PTR_ERR(data); > + > + return sprintf(buf, "%d\n", %u would probably make more sense, as this is used for alarms above. But I agree %d will work too. > + (data->alarms & (1 << 12)) && data->temp[index] == 127); > +} > + > static ssize_t set_temp_max(struct device *dev, > struct device_attribute *devattr, > const char *buf, size_t count) > @@ -276,6 +289,11 @@ static SENSOR_DEVICE_ATTR(temp4_max_alarm, S_IRUGO, show_alarm, NULL, 2); > static SENSOR_DEVICE_ATTR(temp5_min_alarm, S_IRUGO, show_alarm, NULL, 1); > static SENSOR_DEVICE_ATTR(temp5_max_alarm, S_IRUGO, show_alarm, NULL, 0); > > +static SENSOR_DEVICE_ATTR(temp2_fault, S_IRUGO, show_fault, NULL, 1); > +static SENSOR_DEVICE_ATTR(temp3_fault, S_IRUGO, show_fault, NULL, 2); > +static SENSOR_DEVICE_ATTR(temp4_fault, S_IRUGO, show_fault, NULL, 3); > +static SENSOR_DEVICE_ATTR(temp5_fault, S_IRUGO, show_fault, NULL, 4); > + > /* Attributes common to MAX1668, MAX1989 and MAX1805 */ > static struct attribute *max1668_attribute_common[] = { > &sensor_dev_attr_temp1_max.dev_attr.attr, > @@ -294,6 +312,9 @@ static struct attribute *max1668_attribute_common[] = { > &sensor_dev_attr_temp2_min_alarm.dev_attr.attr, > &sensor_dev_attr_temp3_max_alarm.dev_attr.attr, > &sensor_dev_attr_temp3_min_alarm.dev_attr.attr, > + > + &sensor_dev_attr_temp2_fault.dev_attr.attr, > + &sensor_dev_attr_temp3_fault.dev_attr.attr, > NULL > }; > > @@ -310,6 +331,9 @@ static struct attribute *max1668_attribute_unique[] = { > &sensor_dev_attr_temp4_min_alarm.dev_attr.attr, > &sensor_dev_attr_temp5_max_alarm.dev_attr.attr, > &sensor_dev_attr_temp5_min_alarm.dev_attr.attr, > + > + &sensor_dev_attr_temp4_fault.dev_attr.attr, > + &sensor_dev_attr_temp5_fault.dev_attr.attr, > NULL > }; > -- Jean Delvare _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
