diff options
author | Sean Anderson <sean.anderson@linux.dev> | 2024-06-20 17:20:05 -0400 |
---|---|---|
committer | Guenter Roeck <linux@roeck-us.net> | 2024-06-21 08:09:32 -0700 |
commit | 52c1e818d66bfed276bd371f9e7947be4055af87 (patch) | |
tree | d645a9f166f0c85a324470b5960339e52764d980 /drivers/hwmon | |
parent | 1226a1b2e5611d17aae55a01ad9d6883879feab4 (diff) | |
download | lwn-52c1e818d66bfed276bd371f9e7947be4055af87.tar.gz lwn-52c1e818d66bfed276bd371f9e7947be4055af87.zip |
hwmon: iio: Use iio_read_channel_processed_scale for IIO_POWER
Instead of rescaling power channels after the fact, use the dedicated
scaling API. This should reduce any inaccuracies resulting from the
scaling.
Signed-off-by: Sean Anderson <sean.anderson@linux.dev>
Link: https://lore.kernel.org/r/20240620212005.821805-1-sean.anderson@linux.dev
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Diffstat (limited to 'drivers/hwmon')
-rw-r--r-- | drivers/hwmon/iio_hwmon.c | 11 |
1 files changed, 6 insertions, 5 deletions
diff --git a/drivers/hwmon/iio_hwmon.c b/drivers/hwmon/iio_hwmon.c index 4c8a80847891..fab32e1e15f2 100644 --- a/drivers/hwmon/iio_hwmon.c +++ b/drivers/hwmon/iio_hwmon.c @@ -49,16 +49,17 @@ static ssize_t iio_hwmon_read_val(struct device *dev, struct iio_channel *chan = &state->channels[sattr->index]; enum iio_chan_type type; - ret = iio_read_channel_processed(chan, &result); - if (ret < 0) - return ret; - ret = iio_get_channel_type(chan, &type); if (ret < 0) return ret; if (type == IIO_POWER) - result *= 1000; /* mili-Watts to micro-Watts conversion */ + /* mili-Watts to micro-Watts conversion */ + ret = iio_read_channel_processed_scale(chan, &result, 1000); + else + ret = iio_read_channel_processed(chan, &result); + if (ret < 0) + return ret; return sprintf(buf, "%d\n", result); } |