linux-kernel - Re: [PATCH] iio: light: Add raw attribute

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <4e481aa6-eebd-a934-72de-27f0e5e04fb1@collabora.com>
Date:   Thu, 11 Aug 2022 19:48:33 +0530
From:   Shreeya Patel <shreeya.patel@...labora.com>
To:     Jonathan Cameron <Jonathan.Cameron@...wei.com>
Cc:     jic23@...nel.org, lars@...afoo.de, linux-iio@...r.kernel.org,
        linux-kernel@...r.kernel.org, krisman@...labora.com,
        kernel@...labora.com, alvaro.soliverez@...labora.com
Subject: Re: [PATCH] iio: light: Add raw attribute

On 11/08/22 18:55, Jonathan Cameron wrote:
> On Thu, 11 Aug 2022 15:44:43 +0530
> Shreeya Patel <shreeya.patel@...labora.com> wrote:
>
>> Add IIO_CHAN_INFO_RAW to the mask to be able to read raw values
>> from the light sensor.
> Why is this useful?  It's rare to support _PROCESSED and _RAW for the same channel.
> Normally occurred to avoid breaking ABI.

Hi Jonathan,

For some context, we have been using a downstream driver for ltrf216a
in steam deck. The formula for Lux calculation used in that driver is
incorrect which we corrected in the driver that has been upstreamed now.

The userspace code to handle brightness on steam deck uses the 
in_illuminance_input
value from sysfs and multiplies that value with some constant stored in 
BIOS at
factory calibration time.

With the Lux calculation change in upstreamed driver, 
in_illuminance_input gives us
a different value leading to change in the brightness on steam deck. It 
is also not possible
for us to recalculate the constant stored in BIOS.

We need a way to add some magic in userspace code so that brightness 
adjustment
works like before. So I'm trying to calculate a constant using the 
current formula
that we are using :-
( greendata * 45 * LTRF216A_WIN_FAC ) / ( data->als_data_gain * 
data->int_time_fac)

Except for the greendata which is ALS_DATA, I have all the other values 
to calculate
a constant, hence, I added RAW attribute so I can read the ALS_DATA 
value from userspace.

Thanks,
Shreeya Patel

>
> Jonathan
>
>> Signed-off-by: Shreeya Patel <shreeya.patel@...labora.com>
>> ---
>>   drivers/iio/light/ltrf216a.c | 13 +++++++++++++
>>   1 file changed, 13 insertions(+)
>>
>> diff --git a/drivers/iio/light/ltrf216a.c b/drivers/iio/light/ltrf216a.c
>> index e6e24e70d2b9..a717bf99bd2a 100644
>> --- a/drivers/iio/light/ltrf216a.c
>> +++ b/drivers/iio/light/ltrf216a.c
>> @@ -93,6 +93,7 @@ static const struct iio_chan_spec ltrf216a_channels[] = {
>>   	{
>>   		.type = IIO_LIGHT,
>>   		.info_mask_separate =
>> +			BIT(IIO_CHAN_INFO_RAW) |
>>   			BIT(IIO_CHAN_INFO_PROCESSED) |
>>   			BIT(IIO_CHAN_INFO_INT_TIME),
>>   		.info_mask_separate_available =
>> @@ -259,6 +260,18 @@ static int ltrf216a_read_raw(struct iio_dev *indio_dev,
>>   	int ret;
>>   
>>   	switch (mask) {
>> +	case IIO_CHAN_INFO_RAW:
>> +		ret = ltrf216a_set_power_state(data, true);
>> +		if (ret)
>> +			return ret;
>> +		mutex_lock(&data->lock);
>> +		ret = ltrf216a_read_data(data, LTRF216A_ALS_DATA_0);
>> +		mutex_unlock(&data->lock);
>> +		if (ret < 0)
>> +			return ret;
>> +		*val = ret;
>> +		ltrf216a_set_power_state(data, false);
>> +		return IIO_VAL_INT;
>>   	case IIO_CHAN_INFO_PROCESSED:
>>   		mutex_lock(&data->lock);
>>   		ret = ltrf216a_get_lux(data);