All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jonathan Cameron <Jonathan.Cameron@huawei.com>
To: Achim Gratz <Achim.Gratz@Stromeko.DE>
Cc: linux-iio@vger.kernel.org, "Jonathan Cameron" <jic23@kernel.org>,
	"David Lechner" <dlechner@baylibre.com>,
	"Andy Shevchenko" <andy@kernel.org>,
	"Nuno Sá" <nuno.sa@analog.com>
Subject: Re: [bmp280 v1 2/6] iio: pressure: bmp280: reduce overhead on read with MODE_FORCED
Date: Wed, 6 Aug 2025 16:58:59 +0100	[thread overview]
Message-ID: <20250806165859.000039d4@huawei.com> (raw)
In-Reply-To: <20250803140802.36888-3-Achim.Gratz@Stromeko.DE>

On Sun,  3 Aug 2025 16:07:58 +0200
Achim Gratz <Achim.Gratz@Stromeko.DE> wrote:

> When measuring with MODE_FORCED, each read through sysfs triggers a
> new measurement cycle through aLL channels with the current channel
> configuration, even though we can only access a single channel.
> Reduce the incurred overhead (especially for higher oversampling_ratio
> settings) by temporarily switching off the unused channels.  This
> savea about a third of the acquisition time when reading all three
> channels in succession:
> 
> | oversampling | max/full | max/skip | time/full | time/skip |
> |              |     [ms] |     [ms] |      [ms] |      [ms] |
> |--------------+----------+----------+-----------+-----------|
> |           16 |      339 |      120 |       315 |       127 |
> |            8 |      174 |       65 |       166 |        76 |
> |            4 |       90 |       38 |        94 |        53 |
> |            2 |       49 |       24 |        59 |        41 |
> |            1 |       28 |       17 |        36 |        33 |
> 
> The results are from an I²C connected sensor at 400kHz, so there is
> considerable overhead from the changing the channel configuration,
> most noticeably with low oversampling_ratio values.  Faster
> communication will reduce this overhead further; and since there is
> still a net reduction in acquisition time even for
> oversampling_ratio=1 switching off the channels is always done.
> 
> Note: The IIR filters will process a slightly noisier input signal.
> 
> Signed-off-by: Achim Gratz <Achim.Gratz@Stromeko.DE>
Hi Achim.

I'm not really sure what the algorithm implemented here is and what the
various local variables actually mean as state.  Please add some
more comments to the code.

> ---
>  drivers/iio/pressure/bmp280-core.c | 138 ++++++++++++++++++++++++++---
>  drivers/iio/pressure/bmp280.h      |   7 +-
>  2 files changed, 129 insertions(+), 16 deletions(-)
> 
> diff --git a/drivers/iio/pressure/bmp280-core.c b/drivers/iio/pressure/bmp280-core.c
> index 3213dcadba28..858974a64306 100644
> --- a/drivers/iio/pressure/bmp280-core.c
> +++ b/drivers/iio/pressure/bmp280-core.c
> @@ -619,18 +619,66 @@ static int bmp280_read_raw_impl(struct iio_dev *indio_dev,
>  	struct bmp280_data *data = iio_priv(indio_dev);
>  	int chan_value;
>  	int ret;
> +	int prev_oversampling_humid, prev_oversampling_press, prev_oversampling_temp;
> +	int temp_oversampling_humid, temp_oversampling_press, temp_oversampling_temp;
> +	int switch_off, switch_threshold = -1;
>  
>  	guard(mutex)(&data->lock);
>  
> +	prev_oversampling_humid = temp_oversampling_humid = data->oversampling_humid;
> +	prev_oversampling_press = temp_oversampling_press = data->oversampling_press;
> +	prev_oversampling_temp  = temp_oversampling_temp  = data->oversampling_temp;
> +
>  	switch (mask) {
>  	case IIO_CHAN_INFO_PROCESSED:
> +		/* switch off unused channels */
> +		switch_off = 0;

So this is saying if 'anything can be switched off'? 
switch_off_something or some naming like that.

> +		switch (chan->type) {
> +		case IIO_HUMIDITYRELATIVE:
> +			temp_oversampling_press = 0-1;

0 - 1
etc. So always spaces around operators.
I'm not at all sure on the logic behind this maths though.
Seems it ultimately gets store into a u8.  Add some comments on what the algorithm
is doing.


> +			switch_off |= (prev_oversampling_press > switch_threshold);
> +			/* can't be switched off as it is needed for compensation */
What can't be switched off?  The temperature I guess.
> +			temp_oversampling_temp  = 1-1;
1 - 1
> +			break;
> +		case IIO_PRESSURE:
> +			temp_oversampling_humid = 0-1;
> +			switch_off |= (prev_oversampling_humid > switch_threshold);
> +			/* can't be switched off as it is needed for compensation */
> +			temp_oversampling_temp  = 1-1;
> +			break;
> +		case IIO_TEMP:
> +			temp_oversampling_humid = 0-1;
> +			temp_oversampling_press = 0-1;
> +			switch_off = (prev_oversampling_humid > switch_threshold) |
> +				     (prev_oversampling_press > switch_threshold);
> +			break;
> +		default:
> +			return -EINVAL;
> +		}
> +		if (switch_off) {
> +			data->oversampling_humid = temp_oversampling_humid;
> +			data->oversampling_press = temp_oversampling_press;
> +			data->oversampling_temp  = temp_oversampling_temp;
> +			ret = data->chip_info->chip_config(data);
> +			if (ret)
> +				goto restore;
> +		}
> +
>  		ret = data->chip_info->set_mode(data, BMP280_FORCED);
>  		if (ret)
> -			return ret;
> -
> +			goto restore;
>  		ret = data->chip_info->wait_conv(data);
>  		if (ret)
> -			return ret;
> +			goto restore;
> +
> +		if (switch_off) {
> +			data->oversampling_humid = prev_oversampling_humid;
> +			data->oversampling_press = prev_oversampling_press;
> +			data->oversampling_temp  = prev_oversampling_temp;
> +			ret = data->chip_info->chip_config(data);
> +			if (ret)
> +				return ret;
> +		}
>  
>  		switch (chan->type) {
>  		case IIO_HUMIDITYRELATIVE:
> @@ -661,13 +709,55 @@ static int bmp280_read_raw_impl(struct iio_dev *indio_dev,
>  			return -EINVAL;
>  		}
>  	case IIO_CHAN_INFO_RAW:
> +		/* switch off unused channels */
> +		switch_off = 0;
> +		switch (chan->type) {
> +		case IIO_HUMIDITYRELATIVE:
> +			temp_oversampling_press = 0-1;
> +			switch_off |= (prev_oversampling_press > switch_threshold);
> +			/* can't be switched off as it is needed for compensation */
> +			temp_oversampling_temp  = 1-1;
> +			break;
> +		case IIO_PRESSURE:
> +			temp_oversampling_humid = 0-1;
> +			switch_off |= (prev_oversampling_humid > switch_threshold);
> +			/* can't be switched off as it is needed for compensation */
> +			temp_oversampling_temp  = 1-1;
> +			break;
> +		case IIO_TEMP:
> +			temp_oversampling_humid = 0-1;
> +			temp_oversampling_press = 0-1;
> +			switch_off = (prev_oversampling_humid > switch_threshold) |
> +				     (prev_oversampling_press > switch_threshold);
> +			break;
> +		default:
> +			return -EINVAL;
> +		}
> +		if (switch_off) {
> +			data->oversampling_humid = temp_oversampling_humid;
> +			data->oversampling_press = temp_oversampling_press;
> +			data->oversampling_temp  = temp_oversampling_temp;
> +			ret = data->chip_info->chip_config(data);
> +			if (ret)
> +				goto restore;
> +		}
> +
>  		ret = data->chip_info->set_mode(data, BMP280_FORCED);
>  		if (ret)
> -			return ret;
> -
> +			goto restore;
>  		ret = data->chip_info->wait_conv(data);
>  		if (ret)
> -			return ret;
> +			goto restore;
> +
> +		if (switch_off) {
> +			data->oversampling_humid = prev_oversampling_humid;
> +			data->oversampling_press = prev_oversampling_press;
> +			data->oversampling_temp  = prev_oversampling_temp;
> +			data->chip_info->chip_config(data);
> +			if (ret) {
> +				return ret;
> +			}
> +		}
>  
>  		switch (chan->type) {
>  		case IIO_HUMIDITYRELATIVE:
> @@ -741,6 +831,15 @@ static int bmp280_read_raw_impl(struct iio_dev *indio_dev,
>  	default:
>  		return -EINVAL;
>  	}
> +restore:
Maybe worth factoring out he contents of the switch case that has goto restore
into a separate function so that the scope of the restore is same as the code
that is calling it.  (i.e. avoid a goto jumping out of a switch.

> +	if (switch_off) {
> +	/* restore channel configuration */
> +		data->oversampling_humid = prev_oversampling_humid;
> +		data->oversampling_press = prev_oversampling_press;
> +		data->oversampling_temp  = prev_oversampling_temp;
> +		data->chip_info->chip_config(data);
> +	}
> +	return ret;
>  }

  parent reply	other threads:[~2025-08-06 15:59 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-17 23:30 [PATCH v9 0/4] pressure: bmp280: Minor cleanup and interrupt support Vasileios Amoiridis
2024-10-17 23:30 ` [PATCH v9 1/4] iio: pressure: bmp280: Use sleep and forced mode for oneshot captures Vasileios Amoiridis
2025-06-28 18:45   ` ASSI
2025-06-28 20:57     ` David Lechner
2025-06-29  7:43       ` ASSI
2025-07-11 19:17       ` ASSI
2025-07-12 14:49         ` ASSI
2025-07-21 19:15           ` ASSI
2025-07-26 18:34           ` ASSI
2025-07-27 19:08             ` ASSI
2024-10-17 23:30 ` [PATCH v9 2/4] dt-bindings: iio: pressure: bmp085: Add interrupts for BMP3xx and BMP5xx devices Vasileios Amoiridis
2024-10-17 23:30 ` [PATCH v9 3/4] iio: pressure: bmp280: Add data ready trigger support Vasileios Amoiridis
2024-10-17 23:30 ` [PATCH v9 4/4] iio: pressure: bmp280: Move bmp085 interrupt to new configuration Vasileios Amoiridis
2024-10-19 13:55 ` [PATCH v9 0/4] pressure: bmp280: Minor cleanup and interrupt support Jonathan Cameron
2025-08-03 14:07 ` [bmp280 v1 0/6] Fixes and enhancements for the bmp280 driver Achim Gratz
2025-08-03 14:07   ` [bmp280 v1 1/6] iio: pressure: bmp280: correct meas_time_us calculation Achim Gratz
2025-08-06 15:46     ` Jonathan Cameron
2025-08-06 17:53       ` ASSI
2025-08-10 18:04         ` Jonathan Cameron
2025-08-03 14:07   ` [bmp280 v1 2/6] iio: pressure: bmp280: reduce overhead on read with MODE_FORCED Achim Gratz
2025-08-03 20:12     ` Andy Shevchenko
2025-08-06 15:58     ` Jonathan Cameron [this message]
2025-08-06 18:00       ` ASSI
2025-08-03 14:07   ` [bmp280 v1 3/6] iio: pressure: bmp280: implement sampling_frequency for BMx280 Achim Gratz
2025-08-03 20:26     ` Andy Shevchenko
2025-08-04 17:29       ` ASSI
2025-08-10 18:11         ` Jonathan Cameron
2025-08-10 19:12           ` ASSI
2025-08-11 19:48             ` Jonathan Cameron
2025-08-12 19:53               ` ASSI
2025-08-17 15:10                 ` Jonathan Cameron
2025-08-17 16:36                   ` ASSI
2025-08-03 14:08   ` [bmp280 v1 4/6] iio: pressure: bmp280: enable filter settings " Achim Gratz
2025-08-03 20:28     ` Andy Shevchenko
2025-08-04 17:14       ` ASSI
2025-08-10 18:13     ` Jonathan Cameron
2025-08-10 19:01       ` ASSI
2025-08-11 20:14         ` Jonathan Cameron
2025-08-12 19:34           ` ASSI
2025-08-17 14:51             ` Jonathan Cameron
2025-08-03 14:08   ` [bmp280 v1 5/6] iio: pressure: bmp280: remove code duplication Achim Gratz
2025-08-03 20:30     ` Andy Shevchenko
2025-08-10 18:19     ` Jonathan Cameron
2025-08-03 14:08   ` [bmp280 v1 6/6] iio: pressure: bmp280: implement sampling_frequency calculation for BMx280 Achim Gratz
2025-08-03 20:37     ` Andy Shevchenko
2025-08-04 17:20       ` ASSI
2025-08-03 19:20   ` [bmp280 v1 0/6] Fixes and enhancements for the bmp280 driver Andy Shevchenko
2025-08-10 18:58 ` [RFC PATCH v2 0/9] " Achim Gratz
2025-08-10 18:58   ` [RFC PATCH v2 1/9] iio: pressure: bmp280: correct meas_time_us calculation Achim Gratz
2025-08-17 15:16     ` Jonathan Cameron
2025-08-10 18:58   ` [RFC PATCH v2 2/9] iio: pressure: bmp280: implement adaptive wait for BMx280 devices Achim Gratz
2025-08-10 19:49     ` Andy Shevchenko
2025-08-16 18:42       ` ASSI
2025-08-10 18:58   ` [RFC PATCH v2 3/9] iio: pressure: bmp280: implement adaptive wait for BMP380 devices Achim Gratz
2025-08-10 18:58   ` [RFC PATCH v2 4/9] iio: pressure: bmp280: refactoring Achim Gratz
2025-08-17 15:23     ` Jonathan Cameron
2025-08-10 18:58   ` [RFC PATCH v2 5/9] iio: pressure: bmp280: remove code duplication Achim Gratz
2025-08-10 18:58   ` [RFC PATCH v2 6/9] iio: pressure: bmp280: enable filter settings for BMx280 Achim Gratz
2025-08-17 16:37     ` Jonathan Cameron
2025-08-10 18:58   ` [RFC PATCH v2 7/9] iio: pressure: bmp280: implement sampling_frequency " Achim Gratz
2025-08-17 16:53     ` Jonathan Cameron
2025-08-17 17:36       ` ASSI
2025-08-10 18:58   ` [RFC PATCH v2 8/9] iio: pressure: bmp280: implement sampling_frequency calculation " Achim Gratz
2025-08-17 17:04     ` Jonathan Cameron
2025-08-17 17:40       ` ASSI
2025-08-18 17:52         ` Jonathan Cameron
2025-08-10 18:58   ` [RFC PATCH v2 9/9] iio: pressure: bmp280: test longer autosuspend (WIP) Achim Gratz
2025-08-17 17:05     ` Jonathan Cameron
2025-08-17 17:44       ` ASSI
2025-08-11 12:14   ` [RFC PATCH v2 0/9] Fixes and enhancements for the bmp280 driver Andy Shevchenko
2025-08-11 20:17   ` Jonathan Cameron

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20250806165859.000039d4@huawei.com \
    --to=jonathan.cameron@huawei.com \
    --cc=Achim.Gratz@Stromeko.DE \
    --cc=andy@kernel.org \
    --cc=dlechner@baylibre.com \
    --cc=jic23@kernel.org \
    --cc=linux-iio@vger.kernel.org \
    --cc=nuno.sa@analog.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.