iio: processed channel handling fixes + Intel Dollar Cove TI PMIC ADC driver
v4 cover iio: processed channel handling fixes + Intel Dollar Cove TI PMIC ADC driver
Hi All,
Here is v4 of my patch to add an IIO driver for the Intel Dollar Cove TI PMIC ADC. This has turned more into a series with fixes / changes to iio_convert_raw_to_processed() and iio_read_channel_processed_scale(), with the new driver tagged on as the last patch :)
Changes in v4:
- 2 new bug-fixes for iio_convert_raw_to_processed()
- Factor the bugfixed code multiply a s64 and an iio (type, val, val2) triplet out of iio_convert_raw_to_processed() into a new iio_multiply_value() helper
- Add a kunit test for iio_multiply_value()
- Redo the "Improve iio_read_channel_processed_scale() precision" patch using the iio_multiply_value() helper
Changes in v3:
- "iio: Improve iio_read_channel_processed_scale() precision"
- Use div_s64() instead of div_u64() to fix -1.0 - 0.0 range
- Directly return IIO_VAL_INT from valid cases and drop the final return ret after the switch-case
- "iio: adc: Add Intel Dollar Cove TI PMIC ADC driver"
- Use new more compact DC_TI_ADC_DATA_REG_CH(x) macro
- Use regmap_set_bits() regmap_clear_bits() where applicable
- Use regmap_bulk_read() + be16_to_cpu() to read ADC value
- Use sign_extend32() for vbat_zse and vbat_ge reading in probe()
Changes in v2:
- Add new "iio: Improve iio_read_channel_processed_scale() precision" patch to the series
- Add IIO_CHAN_INFO_SCALE and provide ADC scale info for Vbat
- Add IIO_CHAN_INFO_PROCESSED which applies calibration and scaling for the VBat channel
- Address some other small review remarks
Regards,
Hans
Hans de Goede (6):
iio: consumers: Fix handling of negative channel scale in
iio_convert_raw_to_processed()
iio: consumers: Fix offset handling in iio_convert_raw_to_processed()
iio: consumers: Add an iio_multiply_value() helper function
iio: Improve iio_read_channel_processed_scale() precision
iio: test: Add kunit tests for iio_multiply_value()
iio: adc: Add Intel Dollar Cove TI PMIC ADC driver
drivers/iio/adc/Kconfig | 11 + drivers/iio/adc/Makefile | 1 + drivers/iio/adc/intel_dc_ti_adc.c | 327 +++++++++++++++++++++++++++ drivers/iio/inkern.c | 77 ++++--- drivers/iio/test/Kconfig | 12 + drivers/iio/test/Makefile | 1 + drivers/iio/test/iio-test-multiply.c | 209 +++++++++++++++++ include/linux/iio/consumer.h | 18 ++ 8 files changed, 621 insertions(+), 35 deletions(-) create mode 100644 drivers/iio/adc/intel_dc_ti_adc.c create mode 100644 drivers/iio/test/iio-test-multiply.c
Signature · 1 line
2.49.0
PATCH v4 1/6 iio: consumers: Fix handling of negative channel scale in iio_convert_raw_to_processed()
There is an issue with the handling of negative channel scales in iio_convert_raw_to_processed_unlocked() when the channel-scale is of the IIO_VAL_INT_PLUS_[MICRO|NANO] type:
Things work for channel-scale values > -1.0 and < 0.0 because of the use of signed values in:
*processed += div_s64(raw64 * (s64)scale_val2 * scale, 1000000LL);
Things will break however for scale values < -1.0. Lets for example say that raw = 2, (caller-provided)scale = 10 and (channel)scale_val = -1.5.
The result should then be 2 * 10 * -1.5 = -30.
channel-scale = -1.5 means scale_val = -1 and scale_val2 = 500000, now lets see what gets stored in processed:
1. *processed = raw64 * scale_val * scale;
2. *processed += raw64 * scale_val2 * scale / 1000000LL;
1. Sets processed to 2 * -1 * 10 = -20
2. Adds 2 * 500000 * 10 / 1000000 = 10 to processed
And the end result is processed = -20 + 10 = -10, which is not correct.
Fix this by always using the abs value of both scale_val and scale_val2 and if either is negative multiply the end-result by -1.
Note there seems to be an unwritten rule about negative IIO_VAL_INT_PLUS_[MICRO|NANO] values that:
i. values > -1.0 and < 0.0 are written as val=0 val2=-xxx
ii. values <= -1.0 are written as val=-xxx val2=xxx
But iio_format_value() will also correctly display a third option:
iii. values <= -1.0 written as val=-xxx val2=-xxx
Since iio_format_value() uses abs(val) when val2 < 0.
This fix also makes iio_convert_raw_to_processed() properly handle channel-scales using this third option.
---
Changes in v4:
- New patch in v4 of this patch-set
--- drivers/iio/inkern.c | 24 ++++++++++-------------- 1 file changed, 10 insertions(+), 14 deletions(-)
drivers/iio/inkern.c +10 -14
@@ -11,6 +11,7 @@#include <linux/mutex.h> #include <linux/property.h> #include <linux/slab.h> +#include <linux/units.h> #include <linux/iio/iio.h> #include <linux/iio/iio-opaque.h>
@@ -604,7 +605,7 @@ static int iio_convert_raw_to_processed_unlocked(struct iio_channel *chan, {int scale_type, scale_val, scale_val2;int offset_type, offset_val, offset_val2; -s64 raw64 = raw; +s64 denominator, raw64 = raw; offset_type =iio_channel_read (chan, &offset_val, &offset_val2,IIO_CHAN_INFO_OFFSET );
@@ -648,20 +649,15 @@ static int iio_convert_raw_to_processed_unlocked(struct iio_channel *chan, *processed = raw64 * scale_val * scale;break ;case IIO_VAL_INT_PLUS_MICRO : -if (scale_val2 <0 ) - *processed = -raw64 * scale_val * scale; -else - *processed = raw64 * scale_val * scale; - *processed +=div_s64 (raw64 * (s64 )scale_val2 * scale, -1000000LL ); -break ;case IIO_VAL_INT_PLUS_NANO : -if (scale_val2 <0 ) - *processed = -raw64 * scale_val * scale; -else - *processed = raw64 * scale_val * scale; - *processed +=div_s64 (raw64 * (s64 )scale_val2 * scale, -1000000000LL ); +switch (scale_type) { +case IIO_VAL_INT_PLUS_MICRO : denominator =MICRO ;break ; +case IIO_VAL_INT_PLUS_NANO : denominator =NANO ;break ; + } + *processed = raw64 * scale *abs (scale_val); + *processed +=div_s64 (raw64 * scale *abs (scale_val2), denominator); +if (scale_val <0 || scale_val2 <0 ) + *processed *=-1 ;break ;case IIO_VAL_FRACTIONAL : *processed = div_s64(raw64 * (s64)scale_val * scale,
Signature · 1 line
2.49.0
Re: [PATCH v4 1/6] iio: consumers: Fix handling of negative channel scale in iio_convert_raw_to_processed()
Quoted 43 lines · Hans de Goede
There is an issue with the handling of negative channel scales in
iio_convert_raw_to_processed_unlocked()when the channel-scale is of the IIO_VAL_INT_PLUS_[MICRO|NANO] type:Things work for channel-scale values > -1.0 and < 0.0 because of the use of signed values in:
*processed += div_s64(raw64 * (s64)scale_val2 * scale, 1000000LL);Things will break however for scale values < -1.0. Lets for example say that raw = 2, (caller-provided)scale = 10 and (channel)scale_val = -1.5.
The result should then be 2 * 10 * -1.5 = -30.
channel-scale = -1.5 means
scale_val= -1 andscale_val2= 500000, now lets see what gets stored in processed:1. *processed = raw64 *
scale_val* scale;
2. *processed += raw64 *scale_val2* scale / 1000000LL;1. Sets processed to 2 * -1 * 10 = -20
2. Adds 2 * 500000 * 10 / 1000000 = 10 to processedAnd the end result is processed = -20 + 10 = -10, which is not correct.
Fix this by always using the abs value of both
scale_valandscale_val2and if either is negative multiply the end-result by -1.Note there seems to be an unwritten rule about negative IIO_VAL_INT_PLUS_[MICRO|NANO] values that:
i. values > -1.0 and < 0.0 are written as val=0 val2=-xxx
ii. values <= -1.0 are written as val=-xxx val2=xxxBut
iio_format_value()will also correctly display a third option:iii. values <= -1.0 written as val=-xxx val2=-xxx
Since
iio_format_value()uses abs(val) when val2 < 0.This fix also makes
iio_convert_raw_to_processed()properly handle channel-scales using this third option.
...
drivers/iio/inkern.c static int iio_convert_raw_to_processed_unlocked(struct iio_channel *chan, · line 656
Quoted 4 lines
drivers/iio/inkern.c +4
-0+switch (scale_type) { +case IIO_VAL_INT_PLUS_MICRO : denominator =MICRO ;break ; +case IIO_VAL_INT_PLUS_NANO : denominator =NANO ;break ; + }
Now wondering if checkpatch et al. are happy with this style. Not a big deal personally to me, but if we have warnings from the tools perhaps it's better to avoid them.
Signature · 2 lines
With Best Regards, Andy Shevchenko
Re: [PATCH v4 1/6] iio: consumers: Fix handling of negative channel scale in iio_convert_raw_to_processed()
Quoted 47 lines · On Mon, 11 Aug 2025 21:19:36 +0200 Andy Shevchenko <[email protected]>…
Quoted 43 lines · Hans de Goede
There is an issue with the handling of negative channel scales in
iio_convert_raw_to_processed_unlocked()when the channel-scale is of the IIO_VAL_INT_PLUS_[MICRO|NANO] type:Things work for channel-scale values > -1.0 and < 0.0 because of the use of signed values in:
*processed += div_s64(raw64 * (s64)scale_val2 * scale, 1000000LL);Things will break however for scale values < -1.0. Lets for example say that raw = 2, (caller-provided)scale = 10 and (channel)scale_val = -1.5.
The result should then be 2 * 10 * -1.5 = -30.
channel-scale = -1.5 means
scale_val= -1 andscale_val2= 500000, now lets see what gets stored in processed:1. *processed = raw64 *
scale_val* scale;
2. *processed += raw64 *scale_val2* scale / 1000000LL;1. Sets processed to 2 * -1 * 10 = -20
2. Adds 2 * 500000 * 10 / 1000000 = 10 to processedAnd the end result is processed = -20 + 10 = -10, which is not correct.
Fix this by always using the abs value of both
scale_valandscale_val2and if either is negative multiply the end-result by -1.Note there seems to be an unwritten rule about negative IIO_VAL_INT_PLUS_[MICRO|NANO] values that:
i. values > -1.0 and < 0.0 are written as val=0 val2=-xxx
ii. values <= -1.0 are written as val=-xxx val2=xxxBut
iio_format_value()will also correctly display a third option:iii. values <= -1.0 written as val=-xxx val2=-xxx
Since
iio_format_value()uses abs(val) when val2 < 0.This fix also makes
iio_convert_raw_to_processed()properly handle channel-scales using this third option....
Quoted 8 lines
drivers/iio/inkern.c static int iio_convert_raw_to_processed_unlocked(struct iio_channel *chan, · line 656
Quoted 4 lines
drivers/iio/inkern.c +4
-0+switch (scale_type) { +case IIO_VAL_INT_PLUS_MICRO : denominator =MICRO ;break ; +case IIO_VAL_INT_PLUS_NANO : denominator =NANO ;break ; + }Now wondering if checkpatch et al. are happy with this style. Not a big deal personally to me, but if we have warnings from the tools perhaps it's better to avoid them.
Agreed, whilst it's a minor thing let's burn some lines to keep this more standard looking.
case IIO_VAL_INT_PLUS_MICRO: denominator = MICRO; break; case IIO_VAL_INT_PLUS_NANO: denominator = NANO; break; }
Quoted 2 lines
PATCH v4 2/6 iio: consumers: Fix offset handling in iio_convert_raw_to_processed()
In reply to Hans de GoedeFix iio_convert_raw_to_processed() offset handling for channels without a scale attribute.
The offset has been applied to the raw64 value not to the original raw value. Use the raw64 value so that the offset is taken into account.
---
Changes in v4:
- New patch in v4 of this patch-set
--- drivers/iio/inkern.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/inkern.c +1 -1
@@ -640,7 +640,7 @@ static int iio_convert_raw_to_processed_unlocked(struct iio_channel *chan,* If no channel scaling is available apply consumer scale to * raw value and return. */ - *processed = raw * scale; + *processed = raw64 * scale;return 0 ; }
Signature · 1 line
2.49.0
PATCH v4 3/6 iio: consumers: Add an iio_multiply_value() helper function
In reply to Hans de GoedeThe channel-scale handling in iio_convert_raw_to_processed() in essence does the following:
processed = raw * caller-provided-scale * channel-scale
Which can also be written as:
multiplier = raw * caller-provided-scale
iio-val = channel-scale
processed = multiplier * iio-val
Where iioval is a set of IIO_VAL_* type + val + val2 integers, being able to handle multiplication of iio-values like this is something which is useful to have in general and as previous bugfixes to iio_convert_raw_to_processed() have shown also tricky to implement.
Split the iio-value multiplication code from iio_convert_raw_to_processed() out into a new iio_multiply_value() helper. This serves multiple purposes:
1. Having this split out allows writing a kunit test for this.
2. Having this split out allows re-use to get better precision when scaling values in iio_read_channel_processed_scale().
---
Changes in v4:
- New patch in v4 of this patch-set
--- drivers/iio/inkern.c | 64 +++++++++++++++++++++--------------- include/linux/iio/consumer.h | 18 ++++++++++ 2 files changed, 56 insertions(+), 26 deletions(-)
drivers/iio/inkern.c +38 -26
@@ -599,13 +599,46 @@ int iio_read_channel_average_raw(struct iio_channel *chan, int *val) }EXPORT_SYMBOL_GPL (iio_read_channel_average_raw); +int iio_multiply_value (int *result,s64 multiplier, +unsigned int type,int val,int val2) +{ +s64 denominator; + +switch (type) { +case IIO_VAL_INT : + *result = multiplier * val; +return IIO_VAL_INT ; +case IIO_VAL_INT_PLUS_MICRO : +case IIO_VAL_INT_PLUS_NANO : +switch (type) { +case IIO_VAL_INT_PLUS_MICRO : denominator =MICRO ;break ; +case IIO_VAL_INT_PLUS_NANO : denominator =NANO ;break ; + } + *result = multiplier *abs (val); + *result +=div_s64 (multiplier *abs (val2), denominator); +if (val <0 || val2 <0 ) + *result *=-1 ; +return IIO_VAL_INT ; +case IIO_VAL_FRACTIONAL : + *result =div_s64 (multiplier * val, val2); +return IIO_VAL_INT ; +case IIO_VAL_FRACTIONAL_LOG2 : + *result = (multiplier * val) >> val2; +return IIO_VAL_INT ; +default : +return -EINVAL ; + } +} +EXPORT_SYMBOL_GPL (iio_multiply_value); +static int iio_convert_raw_to_processed_unlocked (struct iio_channel *chan,int raw,int *processed,unsigned int scale) {int scale_type, scale_val, scale_val2;int offset_type, offset_val, offset_val2; -s64 denominator, raw64 = raw; +s64 raw64 = raw; +int ret; offset_type =iio_channel_read (chan, &offset_val, &offset_val2,IIO_CHAN_INFO_OFFSET );
@@ -644,31 +677,10 @@ static int iio_convert_raw_to_processed_unlocked(struct iio_channel *chan,return 0 ; } -switch (scale_type) { -case IIO_VAL_INT : - *processed = raw64 * scale_val * scale; -break ; -case IIO_VAL_INT_PLUS_MICRO : -case IIO_VAL_INT_PLUS_NANO : -switch (scale_type) { -case IIO_VAL_INT_PLUS_MICRO : denominator =MICRO ;break ; -case IIO_VAL_INT_PLUS_NANO : denominator =NANO ;break ; - } - *processed = raw64 * scale *abs (scale_val); - *processed +=div_s64 (raw64 * scale *abs (scale_val2), denominator); -if (scale_val <0 || scale_val2 <0 ) - *processed *=-1 ; -break ; -case IIO_VAL_FRACTIONAL : - *processed =div_s64 (raw64 * (s64 )scale_val * scale, - scale_val2); -break ; -case IIO_VAL_FRACTIONAL_LOG2 : - *processed = (raw64 * (s64 )scale_val * scale) >> scale_val2; -break ; -default : -return -EINVAL ; - } + ret =iio_multiply_value (processed, raw64 * scale, + scale_type, scale_val, scale_val2); +if (ret <0 ) +return ret;return 0 ; }
include/linux/iio/consumer.h +18 -0
@@ -381,6 +381,24 @@ int iio_read_channel_offset(struct iio_channel *chan, int *val,int iio_read_channel_scale (struct iio_channel *chan,int *val,int *val2); +/** +* iio_multiply_value() - Multiply an iio value +* @result: Destination pointer for the multiplication result +* @multiplier: Multiplier. +* @type: One of the IIO_VAL_* constants. This decides how the val +* and val2 parameters are interpreted. +* @val: Value being multiplied. +* @val2: Value being multiplied. val2 use depends on type. +* +* Multiply an iio value with a s64 multiplier storing the result as +* IIO_VAL_INT. This is typically used for scaling. +* +* Returns: +* IIO_VAL_INT on success or a negative error-number on failure. +*/ +int iio_multiply_value (int *result,s64 multiplier, +unsigned int type,int val,int val2); +/** * iio_convert_raw_to_processed() - Converts a raw value to a processed value * @chan: The channel being queried
Signature · 1 line
2.49.0
Re: [PATCH v4 3/6] iio: consumers: Add an iio_multiply_value() helper function
Quoted 6 lines · Hans de Goede
The channel-scale handling in
iio_convert_raw_to_processed()in essence does the following:processed = raw * caller-provided-scale * channel-scale
Quoted 7 lines
Which can also be written as:
multiplier = raw * caller-provided-scale
iio-val = channel-scale
processed = multiplier * iio-valWhere iioval is a set of IIO_VAL_* type + val + val2 integers, being
Typo, should be iio-val.
In some cases you use iio-value. Can you make the same things to be spelled in the same way?
Quoted 2 lines
able to handle multiplication of iio-values like this is something which is useful to have in general and as previous bugfixes to
and, as
Quoted 1 line
iio_convert_raw_to_processed()have shown also tricky to implement.
shown. also
Quoted 4 lines
Split the iio-value multiplication code from
iio_convert_raw_to_processed()out into a newiio_multiply_value()helper. This serves multiple purposes:1. Having this split out allows writing a kunit test for this.
KUnit
Quoted 2 lines
2. Having this split out allows re-use to get better precision when scaling values in
iio_read_channel_processed_scale().
...
Quoted 21 lines
drivers/iio/inkern.c +21
-0+int iio_multiply_value (int *result,s64 multiplier, +unsigned int type,int val,int val2) +{ +s64 denominator; + +switch (type) { +case IIO_VAL_INT : + *result = multiplier * val; +return IIO_VAL_INT ; +case IIO_VAL_INT_PLUS_MICRO : +case IIO_VAL_INT_PLUS_NANO : +switch (type) { +case IIO_VAL_INT_PLUS_MICRO : denominator =MICRO ;break ; +case IIO_VAL_INT_PLUS_NANO : denominator =NANO ;break ; + } + *result = multiplier *abs (val); + *result +=div_s64 (multiplier *abs (val2), denominator); +if (val <0 || val2 <0 ) + *result *=-1 ; +return IIO_VAL_INT ; +case IIO_VAL_FRACTIONAL :
drivers/iio/inkern.c int iio_read_channel_average_raw(struct iio_channel *chan, int *val) · line 632
Quoted 10 lines
drivers/iio/inkern.c +10
-0+ *result =div_s64 (multiplier * val, val2); +return IIO_VAL_INT ; +case IIO_VAL_FRACTIONAL_LOG2 : + *result = (multiplier * val) >> val2; +return IIO_VAL_INT ; +default : +return -EINVAL ; + } +} +EXPORT_SYMBOL_GPL (iio_multiply_value);
Is it in the namespace? I think we already discussed with Jonathan that IIO_CORE or something can be started if not yet for this kind of exported symbols (and it doesn't matter if the rest are still using global namespace, we can address that later on).
...
Quoted 2 lines
+/**
+ *iio_multiply_value()- Multiply an iio value
IIO ?
Quoted 3 lines
+ * @result: Destination pointer for the multiplication result
+ * @multiplier: Multiplier.
+ * @type: One of the IIO_VAL_* constants. This decides how the val
@val
Quoted 1 line
+ * and val2 parameters are interpreted.
@val2
Quoted 4 lines
+ * @val: Value being multiplied.
+ * @val2: Value being multiplied. val2 use depends on type.
+ *
+ * Multiply an iio value with a s64 multiplier storing the result as
IIO ?
Quoted 5 lines
+ *
IIO_VAL_INT. This is typically used for scaling.
+ *
+ * Returns:
+ *IIO_VAL_INTon success or a negative error-number on failure.
+ */
Signature · 2 lines
With Best Regards, Andy Shevchenko
Re: [PATCH v4 3/6] iio: consumers: Add an iio_multiply_value() helper function
Quoted 29 lines
Quoted 29 lines
drivers/iio/inkern.c +28
-0+int iio_multiply_value (int *result,s64 multiplier, +unsigned int type,int val,int val2) +{ +s64 denominator; + +switch (type) { +case IIO_VAL_INT : + *result = multiplier * val; +return IIO_VAL_INT ; +case IIO_VAL_INT_PLUS_MICRO : +case IIO_VAL_INT_PLUS_NANO : +switch (type) { +case IIO_VAL_INT_PLUS_MICRO : denominator =MICRO ;break ; +case IIO_VAL_INT_PLUS_NANO : denominator =NANO ;break ; + } + *result = multiplier *abs (val); + *result +=div_s64 (multiplier *abs (val2), denominator); +if (val <0 || val2 <0 ) + *result *=-1 ; +return IIO_VAL_INT ; +case IIO_VAL_FRACTIONAL : + *result =div_s64 (multiplier * val, val2); +return IIO_VAL_INT ; +case IIO_VAL_FRACTIONAL_LOG2 : + *result = (multiplier * val) >> val2; +return IIO_VAL_INT ; +default : +return -EINVAL ;
Quoted 8 lines
drivers/iio/inkern.c int iio_read_channel_average_raw(struct iio_channel *chan, int *val) · line 632
Quoted 3 lines
drivers/iio/inkern.c +3
-0+ } +} +EXPORT_SYMBOL_GPL (iio_multiply_value);Is it in the namespace? I think we already discussed with Jonathan that
IIO_COREor something can be started if not yet for this kind of exported symbols (and it doesn't matter if the rest are still using global namespace, we can address that later on).
I think we are only exporting this for the unit test? If so IIO_UNIT_TEST namespace might be better? We can move it to a more general one if it starts getting use other that for testing (or directly in this file where the export isn't needed).
Quoted 6 lines
...
Quoted 2 lines
+/**
+ *iio_multiply_value()- Multiply an iio value
PATCH v4 4/6 iio: Improve iio_read_channel_processed_scale() precision
In reply to Hans de GoedeBefore this change iio_read_channel_processed_scale() always assumes that channels which advertise IIO_CHAN_INFO_PROCESSED capability return IIO_VAL_INT on success.
Ignoring any fractional values from drivers which return IIO_VAL_INT_PLUS_MICRO / IIO_VAL_INT_PLUS_NANO. These fractional values might become non fractional after scaling so these should be taken into account for better precision.
Use the new iio_multiply_value() helper to do proper scaling taking the fractionional values into account.
---
Changes in v4:
- Use the new iio_multiply_value() helper
--- drivers/iio/inkern.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-)
drivers/iio/inkern.c +3 -4
@@ -722,20 +722,19 @@ int iio_read_channel_processed_scale(struct iio_channel *chan, int *val, unsigned int scale) {struct iio_dev_opaque *iio_dev_opaque =to_iio_dev_opaque (chan->indio_dev); -int ret; +int ret, val2;guard (mutex)(&iio_dev_opaque->info_exist_lock);if (!chan->indio_dev->info)return -ENODEV ;if (iio_channel_has_info (chan->channel,IIO_CHAN_INFO_PROCESSED )) { - ret =iio_channel_read (chan, val,NULL , + ret =iio_channel_read (chan, val, &val2,IIO_CHAN_INFO_PROCESSED );if (ret <0 )return ret; - *val *= scale; -return ret; +return iio_multiply_value (val, scale, ret, *val, val2); }else { ret =iio_channel_read (chan, val,NULL ,IIO_CHAN_INFO_RAW );if (ret <0 )
Signature · 1 line
2.49.0
David Lechner
reviewerRe: [PATCH v4 4/6] iio: Improve iio_read_channel_processed_scale() precision
Quoted 35 lines · Hans de Goede
Before this change
iio_read_channel_processed_scale()always assumes that channels which advertiseIIO_CHAN_INFO_PROCESSEDcapability returnIIO_VAL_INTon success.Ignoring any fractional values from drivers which return
IIO_VAL_INT_PLUS_MICRO/IIO_VAL_INT_PLUS_NANO. These fractional values might become non fractional after scaling so these should be taken into account for better precision.Use the new
iio_multiply_value()helper to do proper scaling taking the fractionional values into account.Signed-off-by Hans de Goede <[email protected]>---
Changes in v4:
- Use the newiio_multiply_value()helper--- drivers/iio/inkern.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-)drivers/iio/inkern.c +1
-1@@ -722,20 +722,19 @@ int iio_read_channel_processed_scale(struct iio_channel *chan, int *val, unsigned int scale) {struct iio_dev_opaque *iio_dev_opaque =to_iio_dev_opaque (chan->indio_dev); -int ret; +int ret, val2;guard (mutex)(&iio_dev_opaque->info_exist_lock);if (!chan->indio_dev->info)return -ENODEV ;
drivers/iio/inkern.c int iio_read_channel_processed_scale(struct iio_channel *chan, int *val, · line 737
Quoted 10 lines
drivers/iio/inkern.c +2
-3if (iio_channel_has_info (chan->channel,IIO_CHAN_INFO_PROCESSED )) { - ret =iio_channel_read (chan, val,NULL , + ret =iio_channel_read (chan, val, &val2,IIO_CHAN_INFO_PROCESSED );if (ret <0 )return ret; - *val *= scale; -return ret; +return iio_multiply_value (val, scale, ret, *val, val2);
Passing val to iio_multiply_value() as 2 different args is a bit confusing.
I would add one more local variable for the processes val. e.g.
int ret, pval, pval2;
Quoted 3 lines
drivers/iio/inkern.c +0
-0}else { ret =iio_channel_read (chan, val,NULL ,IIO_CHAN_INFO_RAW );if (ret <0 )
PATCH v4 5/6 iio: test: Add kunit tests for iio_multiply_value()
In reply to Hans de GoedeAdd kunit tests for iio_multiply_value().
---
Changes in v4:
- New patch in v4 of this patch-set
--- drivers/iio/test/Kconfig | 12 ++ drivers/iio/test/Makefile | 1 + drivers/iio/test/iio-test-multiply.c | 209 +++++++++++++++++++++++++++ 3 files changed, 222 insertions(+) create mode 100644 drivers/iio/test/iio-test-multiply.c
drivers/iio/test/Kconfig +12 -0
@@ -41,3 +41,15 @@ config IIO_FORMAT_KUNIT_TEST to the KUnit documentation in Documentation/dev-tools/kunit/. If unsure, say N. + +config IIO_MULTIPLY_KUNIT_TEST + tristate "Test IIO multiply functions" if !KUNIT_ALL_TESTS + depends on KUNIT + default KUNIT_ALL_TESTS + help + build unit tests for the IIO multiply functions. + + For more information on KUnit and unit tests in general, please refer + to the KUnit documentation in Documentation/dev-tools/kunit/. + + If unsure, say N.
drivers/iio/test/Makefile +1 -0
@@ -7,4 +7,5 @@ obj-$(CONFIG_IIO_RESCALE_KUNIT_TEST) += iio-test-rescale.o obj-$(CONFIG_IIO_FORMAT_KUNIT_TEST) += iio-test-format.o obj-$(CONFIG_IIO_GTS_KUNIT_TEST) += iio-test-gts.o +obj-$(CONFIG_IIO_MULTIPLY_KUNIT_TEST) += iio-test-multiply.o CFLAGS_iio-test-format.o += $(DISABLE_STRUCTLEAK_PLUGIN)
drivers/iio/test/iio-test-multiply.c new file +209 -0
209 lines not on this page · raw message
Signature · 1 line
2.49.0
Re: [PATCH v4 5/6] iio: test: Add kunit tests for iio_multiply_value()
Quoted 2 lines · Hans de Goede
Add kunit tests for
iio_multiply_value().
KUnit
...
Quoted 1 line
+#include <
kunit/test.h>
I believe we still need header inclusions to what we are using outside of the KUnit and IIO consumer scopes, such as types.h, math64.h.
Quoted 1 line
+#include <
linux/iio/consumer.h>
Signature · 2 lines
With Best Regards, Andy Shevchenko
Jonathan Cameron
maintainerRe: [PATCH v4 5/6] iio: test: Add kunit tests for iio_multiply_value()
In reply to Hans de GoedeQuoted 3 lines · On Mon, 11 Aug 2025 17:54:52 +0200 Hans de Goede <[email protected]> wrote:
Add kunit tests for
iio_multiply_value().Signed-off-by Hans de Goede <[email protected]>
Very nice!
I don't have any comments to add to those of other reviewers for any of the series.
PATCH v4 6/6 iio: adc: Add Intel Dollar Cove TI PMIC ADC driver
In reply to Hans de GoedeIntel has 2 completely different "Dollar Cove" PMICs for its Bay Trail / Cherry Trail SoCs. One is made by X-Powers and is called the AXP288. The AXP288's GPADC is already supported by the X-Powers AXP288 ADC driver.
The other "Dollar Cove" PMIC is made by TI and does not have any clear TI denomination, its MFD driver calls it the "Intel Dollar Cove TI PMIC".
Add a driver for the Intel Dollar Cove TI PMIC's general purpose ADC, binding to the "chtdc_ti_adc" MFD cell of the MFD driver.
The "cht" in the cell name comes from Cherry Trail, but it turns out that just like the AXP288 the Intel Dollar Cove TI PMIC is also used with both Intel Bay Trail and Intel Cherry Trail SoCs, so this new ADC driver does not include the cht part in its name.
This is loosely based on kernel/drivers/iio/adc/iio_dc_ti_gpadc.c from the Acer A1-840 Android kernel source-code archive named:
"App. Guide_Acer_20151221_A_A.zip"
which is distributed by Acer from the Acer A1-840 support page:
https://www.acer.com/us-en/support/product-support/A1-840/downloads
---
Changes in v4:
- No changes
Changes in v3:
- Use new more compact DC_TI_ADC_DATA_REG_CH(x) macro
- Use regmap_set_bits(), regmap_clear_bits() where applicable
- Use regmap_bulk_read() + be16_to_cpu() to read ADC value
- Use sign_extend32() for vbat_zse and vbat_ge reading in probe()
Changes in v2:
- Add IIO_CHAN_INFO_SCALE and provide ADC scale info for Vbat
- Add IIO_CHAN_INFO_PROCESSED which applies calibration and scaling for the VBat channel
- Address some other small review remarks
--- drivers/iio/adc/Kconfig | 11 + drivers/iio/adc/Makefile | 1 + drivers/iio/adc/intel_dc_ti_adc.c | 327 ++++++++++++++++++++++++++++++ 3 files changed, 339 insertions(+) create mode 100644 drivers/iio/adc/intel_dc_ti_adc.c
drivers/iio/adc/Kconfig +11 -0
@@ -723,6 +723,17 @@ config INGENIC_ADC This driver can also be built as a module. If so, the module will be called ingenic_adc. +config INTEL_DC_TI_ADC + tristate "Intel Bay / Cherry Trail Dollar Cove TI ADC driver" + depends on INTEL_SOC_PMIC_CHTDC_TI + help + Say yes here to have support for the Dollar Cove TI PMIC ADC device. + Depending on platform configuration, this general purpose ADC can be + used for sensors such as battery voltage and thermal resistors. + + To compile this driver as a module, choose M here: the module will be + called intel_dc_ti_adc. + config INTEL_MRFLD_ADC tristate "Intel Merrifield Basin Cove ADC driver" depends on INTEL_SOC_PMIC_MRFLD
drivers/iio/adc/Makefile +1 -0
@@ -67,6 +67,7 @@ obj-$(CONFIG_IMX8QXP_ADC) += imx8qxp-adc.o obj-$(CONFIG_IMX93_ADC) += imx93_adc.o obj-$(CONFIG_INA2XX_ADC) += ina2xx-adc.o obj-$(CONFIG_INGENIC_ADC) += ingenic-adc.o +obj-$(CONFIG_INTEL_DC_TI_ADC) += intel_dc_ti_adc.o obj-$(CONFIG_INTEL_MRFLD_ADC) += intel_mrfld_adc.o obj-$(CONFIG_LP8788_ADC) += lp8788_adc.o obj-$(CONFIG_LPC18XX_ADC) += lpc18xx_adc.o
drivers/iio/adc/intel_dc_ti_adc.c new file +327 -0
327 lines not on this page · raw message
Signature · 1 line
2.49.0
Re: [PATCH v4 6/6] iio: adc: Add Intel Dollar Cove TI PMIC ADC driver
Quoted 21 lines · Hans de Goede
Intel has 2 completely different "Dollar Cove" PMICs for its Bay Trail / Cherry Trail SoCs. One is made by X-Powers and is called the AXP288. The AXP288's GPADC is already supported by the X-Powers AXP288 ADC driver.
The other "Dollar Cove" PMIC is made by TI and does not have any clear TI denomination, its MFD driver calls it the "Intel Dollar Cove TI PMIC".
Add a driver for the Intel Dollar Cove TI PMIC's general purpose ADC, binding to the "
chtdc_ti_adc" MFD cell of the MFD driver.The "cht" in the cell name comes from Cherry Trail, but it turns out that just like the AXP288 the Intel Dollar Cove TI PMIC is also used with both Intel Bay Trail and Intel Cherry Trail SoCs, so this new ADC driver does not include the cht part in its name.
This is loosely based on
kernel/drivers/iio/adc/iio_dc_ti_gpadc.cfrom the Acer A1-840 Android kernel source-code archive named:
"App. Guide_Acer_20151221_A_A.zip"
which is distributed by Acer from the Acer A1-840 support page:
https://www.acer.com/us-en/support/product-support/A1-840/downloads
...
drivers/iio/adc/Kconfig config INGENIC_ADC · line 727
Quoted 2 lines
drivers/iio/adc/Kconfig +2
-0+config INTEL_DC_TI_ADC + tristate "Intel Bay / Cherry Trail Dollar Cove TI ADC driver"
I would spell it in full
"Intel Bay Trail / Cherry Trail ..."
Quoted 1 line
+ depends on
INTEL_SOC_PMIC_CHTDC_TI
...
drivers/iio/adc/intel_dc_ti_adc.c line 157
Quoted 3 lines
drivers/iio/adc/intel_dc_ti_adc.c new file +3
-0+/* +* As per TI (PMIC Vendor), the ADC enable and ADC start commands should +* not be sent together. Hence send the commands separately
Missing period.
Quoted 1 line
+ */
...
Review
Quoted 2 lines
(quoted diff) +2
-0+ /* + * As per PMIC Vendor, a minimum of 50 micro seconds delay is required
We can also use SI unit acronyms, like µs.
Quoted 6 lines
drivers/iio/adc/intel_dc_ti_adc.c new file +6
-0+* between ADC Enable and ADC START commands. This is also recommended +* by Intel Hardware team after the timing analysis of GPADC signals. +* Since the I2C Write transaction to set the channel number also +* imparts 25 micro seconds of delay, so we need to wait for another +* 25 micro seconds before issuing ADC START command. +*/
...
Quoted 1 line
+ /* TI (PMIC Vendor) recommends 5 sec timeout for conversion */
seconds (to be consistent with other comments) or s (to be consistent with SI). Interestingly, you have all three variants among the comments already.
...
Quoted 1 line
+ /* The ADC values are 10 bits */
I would write it as "...are 10-bit wide"
...
drivers/iio/adc/intel_dc_ti_adc.c line 221
Quoted 7 lines
drivers/iio/adc/intel_dc_ti_adc.c new file +6
-0+/* +* If channel BPTHERM has been selected, first enable the BPTHERM BIAS +* which provides the VREF Voltage reference to convert BPTHERM Input +* voltage to temperature. +* As per PMIC Vendor specifications, BPTHERM BIAS should be enabled +* 35 ms before ADC_EN command.
I would move this comment to be directly before msleep() call.
Quoted 63 lines
drivers/iio/adc/intel_dc_ti_adc.c new file +62
-0+*/ +if (chan->channel ==DC_TI_ADC_BATTEMP ) { + ret =regmap_set_bits (info->regmap,DC_TI_ADC_CNTL_REG , +DC_TI_ADC_EN_EXT_BPTH_BIAS ); +if (ret) +return ret; +msleep (35 ); + } +static int dc_ti_adc_probe (struct platform_device *pdev) +{ +struct device *dev = &pdev->dev; +struct intel_soc_pmic *pmic =dev_get_drvdata (dev->parent); +struct dc_ti_adc_info *info; +struct iio_dev *indio_dev; +unsigned int val; +int irq, ret; + + irq =platform_get_irq (pdev,0 ); +if (irq <0 ) +return irq; + + indio_dev =devm_iio_device_alloc (dev,sizeof (*info)); +if (!indio_dev) +return -ENOMEM ; + + info =iio_priv (indio_dev); + + ret =devm_mutex_init (dev, &info->lock); +if (ret) +return ret; + +init_waitqueue_head (&info->wait); + + info->dev = dev; + info->regmap = pmic->regmap; + + indio_dev->name ="dc_ti_adc" ; + indio_dev->channels = dc_ti_adc_channels; + indio_dev->num_channels =ARRAY_SIZE (dc_ti_adc_channels); + indio_dev->info = &dc_ti_adc_iio_info; + indio_dev->modes =INDIO_DIRECT_MODE ; + + ret =regmap_read (info->regmap,DC_TI_VBAT_ZSE_GE_REG , &val); +if (ret) +return ret; + + info->vbat_zse =sign_extend32 (FIELD_GET (DC_TI_VBAT_ZSE , val),3 ); + info->vbat_ge =sign_extend32 (FIELD_GET (DC_TI_VBAT_GE , val),3 ); + +dev_dbg (dev,"vbat-zse %d vbat-ge %d\n" , info->vbat_zse, info->vbat_ge); + + ret =devm_iio_map_array_register (dev, indio_dev, dc_ti_adc_default_maps); +if (ret) +return ret; + + ret =devm_request_threaded_irq (dev, irq,NULL , dc_ti_adc_isr, +IRQF_ONESHOT , indio_dev->name, info); +if (ret) +return ret; + +return devm_iio_device_register (dev, indio_dev); +}
...
drivers/iio/adc/intel_dc_ti_adc.c line 310
Quoted 2 lines
drivers/iio/adc/intel_dc_ti_adc.c new file +2
-0+static const struct platform_device_id dc_ti_adc_ids[] = { + { .name ="chtdc_ti_adc" , },
Inner comma is not required.
Quoted 2 lines
(quoted diff) +2
-0+ { } +};
...
Quoted 1 line
+MODULE_AUTHOR("Ramakrishna Pallala <[email protected]>");
Wondering if the address is still alive. In this case perhaps even ask for SoB/Co-developed-by? Otherwise if you still wish to have this credit make it in the form of "First Last (Intel)".
Signature · 2 lines
With Best Regards, Andy Shevchenko
Re: [PATCH v4 6/6] iio: adc: Add Intel Dollar Cove TI PMIC ADC driver
Hi Andy,
Quoted 17 lines · Andy Shevchenko
Quoted 16 lines · Hans de Goede
Intel has 2 completely different "Dollar Cove" PMICs for its Bay Trail / Cherry Trail SoCs. One is made by X-Powers and is called the AXP288. The AXP288's GPADC is already supported by the X-Powers AXP288 ADC driver.
The other "Dollar Cove" PMIC is made by TI and does not have any clear TI denomination, its MFD driver calls it the "Intel Dollar Cove TI PMIC".
Add a driver for the Intel Dollar Cove TI PMIC's general purpose ADC, binding to the "
chtdc_ti_adc" MFD cell of the MFD driver.The "cht" in the cell name comes from Cherry Trail, but it turns out that just like the AXP288 the Intel Dollar Cove TI PMIC is also used with both Intel Bay Trail and Intel Cherry Trail SoCs, so this new ADC driver does not include the cht part in its name.
Quoted 5 lines
Quoted 5 lines
This is loosely based on
kernel/drivers/iio/adc/iio_dc_ti_gpadc.cfrom the Acer A1-840 Android kernel source-code archive named:
"App. Guide_Acer_20151221_A_A.zip"
which is distributed by Acer from the Acer A1-840 support page:
https://www.acer.com/us-en/support/product-support/A1-840/downloads
Thank you for the review. I've addressed all your remarks for v5.
...
Quoted 5 lines
Quoted 1 line
+MODULE_AUTHOR("Ramakrishna Pallala <[email protected]>");
Wondering if the address is still alive. In this case perhaps even ask for SoB/Co-developed-by? Otherwise if you still wish to have this credit make it in the form of "First Last (Intel)".
Good point, I asked about co-maintaining and got a bounce so I'll replace this with "First Last (Intel)".
Regards,
Hans
Quoted 5 lines · Hans de Goede
Here is v4 of my patch to add an IIO driver for the Intel Dollar Cove TI PMIC ADC. This has turned more into a series with fixes / changes to
iio_convert_raw_to_processed()andiio_read_channel_processed_scale(), with the new driver tagged on as the last patch :)
Several nit-picks here and there, but in general I have no objections.
Signature · 2 lines
With Best Regards, Andy Shevchenko
Hi All,
Quoted 6 lines · Hans de Goede
Hi All,
Here is v4 of my patch to add an IIO driver for the Intel Dollar Cove TI PMIC ADC. This has turned more into a series with fixes / changes to
iio_convert_raw_to_processed()andiio_read_channel_processed_scale(), with the new driver tagged on as the last patch :)
Thank you for all the review comments.
I was away for vacation and I'm catching up on email now.
I agree with all the review comments and I'll prepare a v5 addressing the review comments when I can make some time for this.
Regards,
Hans
Quoted 54 lines
Changes in v4:
- 2 new bug-fixes foriio_convert_raw_to_processed()
- Factor the bugfixed code multiply a s64 and an iio (type, val, val2) triplet out ofiio_convert_raw_to_processed()into a newiio_multiply_value()helper
- Add a kunit test foriio_multiply_value()
- Redo the "Improveiio_read_channel_processed_scale()precision" patch using theiio_multiply_value()helperChanges in v3:
- "iio: Improveiio_read_channel_processed_scale()precision"
- Usediv_s64()instead ofdiv_u64()to fix -1.0 - 0.0 range
- Directly returnIIO_VAL_INTfrom valid cases and drop the final return ret after the switch-case
- "iio: adc: Add Intel Dollar Cove TI PMIC ADC driver"
- Use new more compactDC_TI_ADC_DATA_REG_CH(x)macro
- Useregmap_set_bits()regmap_clear_bits()where applicable
- Useregmap_bulk_read()+be16_to_cpu()to read ADC value
- Usesign_extend32()forvbat_zseandvbat_gereading inprobe()Changes in v2:
- Add new "iio: Improveiio_read_channel_processed_scale()precision" patch to the series
- AddIIO_CHAN_INFO_SCALEand provide ADC scale info for Vbat
- AddIIO_CHAN_INFO_PROCESSEDwhich applies calibration and scaling for the VBat channel
- Address some other small review remarksRegards,
Hans
Hans de Goede (6):
iio: consumers: Fix handling of negative channel scale in
iio_convert_raw_to_processed()
iio: consumers: Fix offset handling iniio_convert_raw_to_processed()
iio: consumers: Add aniio_multiply_value()helper function
iio: Improveiio_read_channel_processed_scale()precision
iio: test: Add kunit tests foriio_multiply_value()
iio: adc: Add Intel Dollar Cove TI PMIC ADC driverdrivers/iio/adc/Kconfig | 11 + drivers/iio/adc/Makefile | 1 + drivers/iio/adc/intel_dc_ti_adc.c | 327 +++++++++++++++++++++++++++ drivers/iio/inkern.c | 77 ++++--- drivers/iio/test/Kconfig | 12 + drivers/iio/test/Makefile | 1 + drivers/iio/test/iio-test-multiply.c | 209 +++++++++++++++++ include/linux/iio/consumer.h | 18 ++ 8 files changed, 621 insertions(+), 35 deletions(-) create mode 100644 drivers/iio/adc/intel_dc_ti_adc.c create mode 100644 drivers/iio/test/iio-test-multiply.c