Re: [PATCH 2/3] hid-sensor-common: Add relative sensitivity check

From: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
To: Jonathan Cameron <jic23@kernel.org>, Ye Xiang <xiang.ye@intel.com>
Cc: jikos@kernel.org, linux-input@vger.kernel.org,
	linux-iio@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/3] hid-sensor-common: Add relative sensitivity check
Date: Sun, 24 Jan 2021 08:20:12 -0800	[thread overview]
Message-ID: <7e136ebb914f71da3fcb90b8048f9f7dd8cdf0bf.camel@linux.intel.com> (raw)
In-Reply-To: <20210124131442.0fc2577e@archlinux>

On Sun, 2021-01-24 at 13:14 +0000, Jonathan Cameron wrote:
> On Wed, 20 Jan 2021 15:47:05 +0800
> Ye Xiang <xiang.ye@intel.com> wrote:
> 
> > Some hid sensors may use relative sensitivity such as als sensor.
> > This patch add relative sensitivity check for all hid-sensors.
> > 
> > Signed-off-by: Ye Xiang <xiang.ye@intel.com>
> > ---
> >  .../iio/common/hid-sensors/hid-sensor-attributes.c    | 11
> > ++++++++++-
> >  include/linux/hid-sensor-ids.h                        |  1 +
> >  2 files changed, 11 insertions(+), 1 deletion(-)
> > 
> > diff --git a/drivers/iio/common/hid-sensors/hid-sensor-attributes.c 
> > b/drivers/iio/common/hid-sensors/hid-sensor-attributes.c
> > index d349ace2e33f..b685c292a179 100644
> > --- a/drivers/iio/common/hid-sensors/hid-sensor-attributes.c
> > +++ b/drivers/iio/common/hid-sensors/hid-sensor-attributes.c
> > @@ -480,7 +480,7 @@ int hid_sensor_parse_common_attributes(struct
> > hid_sensor_hub_device *hsdev,
> >  
> >  	/*
> >  	 * Set Sensitivity field ids, when there is no individual
> > modifier, will
> > -	 * check absolute sensitivity of data field
> > +	 * check absolute sensitivity and relative sensitivity of data
> > field
> >  	 */
> >  	for (i = 0; i < sensitivity_addresses_len && st-
> > >sensitivity.index < 0; i++) {
> >  		sensor_hub_input_get_attribute_info(hsdev,
> > @@ -488,6 +488,15 @@ int hid_sensor_parse_common_attributes(struct
> > hid_sensor_hub_device *hsdev,
> >  				HID_USAGE_SENSOR_DATA_MOD_CHANGE_SENSIT
> > IVITY_ABS |
> >  					sensitivity_addresses[i],
> >  				&st->sensitivity);
> > +
> > +		if (st->sensitivity.index >= 0)
> > +			break;
> > +
> > +		sensor_hub_input_get_attribute_info(hsdev,
> > +				HID_FEATURE_REPORT, usage_id,
> > +				HID_USAGE_SENSOR_DATA_MOD_CHANGE_SENSIT
> > IVITY_REL_PCT |
> > +					sensitivity_addresses[i],
> > +				&st->sensitivity);
> 
> We can't provide the value to userspace without reflecting the
> difference between
> the two ways of expressing it.
> 
> It seems there are 3 ways sensitivity is expressed.
> 1. Raw value in same units as the measurement (easy one and what is
> currently reported)
> 2. Percentage of range - also relatively easy to transform into the
> same as 1.
> 3. Percentage of prior reading..  This one doesn't fit in any
> existing ABI, so
>    unfortunately we'll have to invent something new along the lines
> of
>    *_hysteresis_relative 

This is why it was not added before when I developed.  But later few
years back there was a patch to add this by one of our developer. There
was some discussion, I thought it was decided it is OK to add.

But I agree, we should add new ABI as you suggested. Now almost every
laptop has HID sensors, better to address this. 

Thanks,
Srinivas

> Jonathan
> 
> 
> 
> >  	}
> >  
> >  	st->raw_hystersis = -1;
> > diff --git a/include/linux/hid-sensor-ids.h b/include/linux/hid-
> > sensor-ids.h
> > index 3bbdbccc5805..ac631159403a 100644
> > --- a/include/linux/hid-sensor-ids.h
> > +++ b/include/linux/hid-sensor-ids.h
> > @@ -149,6 +149,7 @@
> >  /* Per data field properties */
> >  #define HID_USAGE_SENSOR_DATA_MOD_NONE				
> > 	0x00
> >  #define HID_USAGE_SENSOR_DATA_MOD_CHANGE_SENSITIVITY_ABS		
> > 0x1000
> > +#define
> > HID_USAGE_SENSOR_DATA_MOD_CHANGE_SENSITIVITY_REL_PCT            0xE
> > 000
> >  
> >  /* Power state enumerations */
> >  #define HID_USAGE_SENSOR_PROP_POWER_STATE_UNDEFINED_ENUM	0x20085
> > 0