Skip to contentSkip to conversation

iio: pressure: bmp280: Small driver cleanup

PATCH In review Angel Iglesias 3 participants 10 messages Aug 13, 2023 – Sep 7, 2023 IIO

lore
Share & tools
lore
#0 lore raw

v2 cover iio: pressure: bmp280: Small driver cleanup

Minor cleanup of BMP280 i2c code and migration to the new helper functions i2c_get_match_data() and spi_get_device_match_data().

Revise driver coding style to follow reverse christmas tree.

Changelog v2:
- Added patch 3 adopting spi_get_device_match_data().
- Updated patch 2 following advice from Uwe Kleine-König <[email protected]> regarding unintentioned driver semantic changes.
- Revised thoroughfully driver coding style and update more functions.

v1:
https://lore.kernel.org/all/[email protected]/

Angel Iglesias (3):
iio: pressure: bmp280: i2c: Rearrange vars in reverse xmas tree order
iio: pressure: bmp280: Use i2c_get_match_data
iio: pressure: bmp280: Use spi_get_device_match_data()

 drivers/iio/pressure/bmp280-core.c |  4 ++--
 drivers/iio/pressure/bmp280-i2c.c  |  8 +++-----
 drivers/iio/pressure/bmp280-spi.c  | 10 +++-------
 3 files changed, 8 insertions(+), 14 deletions(-)

base-commit: 14b7447cec15ee8dfdfe0da66ba1e280ded7e00a

Signature · 1 line
2.41.0
#1 lore raw

PATCH v2 1/3 iio: pressure: bmp280: i2c: Rearrange vars in reverse xmas tree order

Minor cleanup reordering local variable declarations following reverse christmas tree convention.

drivers/iio/pressure/bmp280-core.c +2 -2
index 6089f3f9d8f4..ea02a623bb58 100644
@@ -766,8 +766,8 @@ static const struct iio_info bmp280_info = {
 
 static int bmp280_chip_config(struct bmp280_data *data)
 {
-	u8 osrs = FIELD_PREP(BMP280_OSRS_TEMP_MASK, data->oversampling_temp + 1) |
-		  FIELD_PREP(BMP280_OSRS_PRESS_MASK, data->oversampling_press + 1);
+	u8 osrs = FIELD_PREP(BMP280_OSRS_PRESS_MASK, data->oversampling_press + 1) |
+		  FIELD_PREP(BMP280_OSRS_TEMP_MASK, data->oversampling_temp + 1);
 	int ret;
 
 	ret = regmap_write_bits(data->regmap, BMP280_REG_CTRL_MEAS,
drivers/iio/pressure/bmp280-i2c.c +2 -2
index dbe630ad05b5..693eb1975fdc 100644
@@ -7,9 +7,9 @@
 
 static int bmp280_i2c_probe(struct i2c_client *client)
 {
-	struct regmap *regmap;
-	const struct bmp280_chip_info *chip_info;
 	const struct i2c_device_id *id = i2c_client_get_device_id(client);
+	const struct bmp280_chip_info *chip_info;
+	struct regmap *regmap;
 
 	chip_info = device_get_match_data(&client->dev);
 	if (!chip_info)
drivers/iio/pressure/bmp280-spi.c +2 -4
index 1dff9bb7c4e9..1c9c01f1b1c7 100644
@@ -14,8 +14,7 @@
 static int bmp280_regmap_spi_write(void *context, const void *data,
                                    size_t count)
 {
-	struct device *dev = context;
-	struct spi_device *spi = to_spi_device(dev);
+	struct spi_device *spi = to_spi_device(context);
 	u8 buf[2];
 
 	memcpy(buf, data, 2);
@@ -31,8 +30,7 @@ static int bmp280_regmap_spi_write(void *context, const void *data,
 static int bmp280_regmap_spi_read(void *context, const void *reg,
                                   size_t reg_size, void *val, size_t val_size)
 {
-	struct device *dev = context;
-	struct spi_device *spi = to_spi_device(dev);
+	struct spi_device *spi = to_spi_device(context);
 
 	return spi_write_then_read(spi, reg, reg_size, val, val_size);
 }
Signature · 1 line
2.41.0
#2 lore raw

PATCH v2 2/3 iio: pressure: bmp280: Use i2c_get_match_data

In reply to Angel Iglesias

Replaces device_get_match_data() and fallback match_id logic by new unified helper function i2c_get_match_data().

drivers/iio/pressure/bmp280-i2c.c +1 -3
index 693eb1975fdc..34e3bc758493 100644
@@ -11,9 +11,7 @@ static int bmp280_i2c_probe(struct i2c_client *client)
 	const struct bmp280_chip_info *chip_info;
 	struct regmap *regmap;
 
-	chip_info = device_get_match_data(&client->dev);
-	if (!chip_info)
-		chip_info = (const struct bmp280_chip_info *) id->driver_data;
+	chip_info = i2c_get_match_data(client);
 
 	regmap = devm_regmap_init_i2c(client, chip_info->regmap_config);
 	if (IS_ERR(regmap)) {
Signature · 1 line
2.41.0
#9 lore raw

PATCH v2 3/3 iio: pressure: bmp280: Use spi_get_device_match_data()

In reply to Angel Iglesias

Use the spi_get_device_match_data() helper instead of device_get_match_data() and the fallback match_id logic.

drivers/iio/pressure/bmp280-spi.c +1 -3
index 1c9c01f1b1c7..433d6fac83c4 100644
@@ -56,9 +56,7 @@ static int bmp280_spi_probe(struct spi_device *spi)
 		return ret;
 	}
 
-	chip_info = device_get_match_data(&spi->dev);
-	if (!chip_info)
-		chip_info = (const struct bmp280_chip_info *) id->driver_data;
+	chip_info = spi_get_device_match_data(spi);
 
 	regmap = devm_regmap_init(&spi->dev,
 				  &bmp280_regmap_bus,
Signature · 1 line
2.41.0
#3 lore raw

Re: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

In reply to Angel Iglesias
Quoted 13 lines · Angel Iglesias

Replaces device_get_match_data() and fallback match_id logic by new unified helper function i2c_get_match_data().

drivers/iio/pressure/bmp280- renamed from drivers/iio/pressure/bmp280-i2c.c +0 -0

i2c.c
index 693eb1975fdc..34e3bc758493 100644

drivers/iio/pressure/bmp280-i2c.c +0 -0
@@ -11,9 +11,7 @@ static int bmp280_i2c_probe(struct i2c_client *client)
         const struct bmp280_chip_info *chip_info;
         struct regmap *regmap;
Quoted 8 lines


-       chip_info = device_get_match_data(&client->dev);
-       if (!chip_info)
-               chip_info = (const struct bmp280_chip_info *) id->driver_data; +       chip_info = i2c_get_match_data(client);

        regmap = devm_regmap_init_i2c(client, chip_info->regmap_config); if (IS_ERR(regmap)) {

Hi,

I noticed I submitted this change that was also submitted by Biju Das on another patch:
https://lore.kernel.org/all/[email protected]/

Should I drop this patch from the series?

Kind regards,
Angel

#4 lore raw

RE: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

Hi Angel Iglesias,

Quoted 30 lines

Subject: Re: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

Quoted 21 lines · Angel Iglesias

Replaces device_get_match_data() and fallback match_id logic by new unified helper function i2c_get_match_data().

(quoted diff) +0 -0

b/drivers/iio/pressure/bmp280- i2c.c index 693eb1975fdc..34e3bc758493 100644

drivers/iio/pressure/bmp280-i2c.c +1 -3
@@ -11,9 +11,7 @@ static int bmp280_i2c_probe(struct i2c_client
 *client)
         const struct bmp280_chip_info *chip_info;
         struct regmap *regmap;
 
-       chip_info = device_get_match_data(&client->dev);
-       if (!chip_info)
-               chip_info = (const struct bmp280_chip_info *)
 id->driver_data;
+       chip_info = i2c_get_match_data(client);
 
Quoted 3 lines

        regmap = devm_regmap_init_i2c(client,
chip_info->regmap_config);
        if (IS_ERR(regmap)) {

Hi,

Quoted 5 lines

I noticed I submitted this change that was also submitted by Biju Das on another
patch:

Should I drop this patch from the series?

I think it is ok. Andy is suggesting to use unified table for SPI/I2C

Is it something do able and testable in your environment? see [1], If yes, please create another patch for using unified table for both i2c and spi.

https://lore.kernel.org/linux-renesas-soc/CAHp75VeX+T=hAN+PgtHTdv4b6UtDVgveUUww1b1kuOngzDinFw@mail.gmail.com/T/#t

Cheers,
Biju

#5 lore raw

Re: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

Quoted 37 lines · Biju Das

Hi Angel Iglesias,

Quoted 33 lines

Subject: Re: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

Quoted 21 lines · Angel Iglesias

Replaces device_get_match_data() and fallback match_id logic by new unified helper function i2c_get_match_data().

(quoted diff) +0 -0

b/drivers/iio/pressure/bmp280- i2c.c index 693eb1975fdc..34e3bc758493 100644

drivers/iio/pressure/bmp280-i2c.c +1 -3
@@ -11,9 +11,7 @@ static int bmp280_i2c_probe(struct i2c_client
 *client)
         const struct bmp280_chip_info *chip_info;
         struct regmap *regmap;
 
-       chip_info = device_get_match_data(&client->dev);
-       if (!chip_info)
-               chip_info = (const struct bmp280_chip_info *)
 id->driver_data;
+       chip_info = i2c_get_match_data(client);
 
Quoted 3 lines

        regmap = devm_regmap_init_i2c(client,
chip_info->regmap_config);
        if (IS_ERR(regmap)) {

Hi,

I noticed I submitted this change that was also submitted by Biju Das on another
patch:

Quoted 7 lines
Quoted 1 line

Should I drop this patch from the series?

I think it is ok. Andy is suggesting to use unified table for SPI/I2C

Is it something do able and testable in your environment? see [1], If yes, please create another patch for using unified table for both i2c and spi.

I have around a BMP390 with the SPI pins available to test it out. In the case of the bmp280 we could unify the of_match table as they're almost the same. In the case of the spi_device_id and i2c_device_id tables, as they're different structs I'm not sure if they can be unified.

Regarding Andy's comment, I think he's referring to the duplicated chip infos. In the case of the bmp280, the chip_infos are defined on the common driver code and used for both SPI and I2C match tables.

Quoted 5 lines

https://lore.kernel.org/linux-renesas-soc/CAHp75VeX+T=hAN+PgtHTdv4b6UtDVgveUUww1b1kuOngzDinFw@mail.gmail.com/T/#t

Cheers,
Biju

Kind regards,
Angel

maintainer
#6 lore raw

Re: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

Quoted 46 lines · On Mon, 14 Aug 2023 18:43:49 +0200 Angel Iglesias <[email protected]>…
Quoted 37 lines · Biju Das

Hi Angel Iglesias,

Quoted 33 lines

Subject: Re: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

Quoted 21 lines · Angel Iglesias

Replaces device_get_match_data() and fallback match_id logic by new unified helper function i2c_get_match_data().

(quoted diff) +0 -0

b/drivers/iio/pressure/bmp280- i2c.c index 693eb1975fdc..34e3bc758493 100644

drivers/iio/pressure/bmp280-i2c.c +1 -3
@@ -11,9 +11,7 @@ static int bmp280_i2c_probe(struct i2c_client
 *client)
         const struct bmp280_chip_info *chip_info;
         struct regmap *regmap;
 
-       chip_info = device_get_match_data(&client->dev);
-       if (!chip_info)
-               chip_info = (const struct bmp280_chip_info *)
 id->driver_data;
+       chip_info = i2c_get_match_data(client);
 
Quoted 3 lines

        regmap = devm_regmap_init_i2c(client,
chip_info->regmap_config);
        if (IS_ERR(regmap)) {

Hi,

I noticed I submitted this change that was also submitted by Biju Das on another
patch:

Quoted 7 lines
Quoted 1 line

Should I drop this patch from the series?

I think it is ok. Andy is suggesting to use unified table for SPI/I2C

Is it something do able and testable in your environment? see [1], If yes, please create another patch for using unified table for both i2c and spi.

Quoted 8 lines

I have around a BMP390 with the SPI pins available to test it out. In the case of the bmp280 we could unify the of_match table as they're almost the same. In the case of the spi_device_id and i2c_device_id tables, as they're different structs I'm not sure if they can be unified.

Regarding Andy's comment, I think he's referring to the duplicated chip infos. In the case of the bmp280, the chip_infos are defined on the common driver code and used for both SPI and I2C match tables.

Hi,

I'm loosing track of where we are with this driver as multiple people are working on it.

Angel, as most of the work is yours, please could you manage the flow of patches for this one so I get series with clear statement of what they are dependent on.

Thanks,

Jonathan

Quoted 10 lines
Quoted 6 lines

https://lore.kernel.org/linux-renesas-soc/CAHp75VeX+T=hAN+PgtHTdv4b6UtDVgveUUww1b1kuOngzDinFw@mail.gmail.com/T/#t

Cheers,
Biju

Kind regards,
Angel

#8 lore raw

RE: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

In reply to Angel Iglesias

Hi Angel Iglesias,

Quoted 72 lines

Subject: Re: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

Quoted 69 lines · Jonathan Cameron
Quoted 57 lines · On Mon, 14 Aug 2023 18:43:49 +0200 Angel Iglesias <[email protected]>…
Quoted 40 lines · Biju Das

Hi Angel Iglesias,

Quoted 36 lines

Subject: Re: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

Quoted 23 lines · Angel Iglesias

Replaces device_get_match_data() and fallback match_id logic by new unified helper function i2c_get_match_data().

(quoted diff) +0 -0

b/drivers/iio/pressure/bmp280- i2c.c index
693eb1975fdc..34e3bc758493
100644

drivers/iio/pressure/bmp280-i2c.c +0 -0
@@ -11,9 +11,7 @@ static int bmp280_i2c_probe(struct

i2c_client
*client)
        const struct bmp280_chip_info *chip_info;
        struct regmap *regmap;

-       chip_info = device_get_match_data(&client->dev); -       if (!chip_info)
-               chip_info = (const struct bmp280_chip_info *) id->driver_data;
+       chip_info = i2c_get_match_data(client);

Quoted 3 lines

        regmap = devm_regmap_init_i2c(client,
chip_info->regmap_config);
        if (IS_ERR(regmap)) {

Hi,

I noticed I submitted this change that was also submitted by Biju Das on another
patch:

Quoted 8 lines
Quoted 1 line

Should I drop this patch from the series?

I think it is ok. Andy is suggesting to use unified table for SPI/I2C

Is it something do able and testable in your environment? see [1], If yes, please create another patch for using unified table for both i2c and spi.

I have around a BMP390 with the SPI pins available to test it out. In the case of the bmp280 we could unify the of_match table as they're almost the same.
In
the case of the spi_device_id and i2c_device_id tables, as they're different structs I'm not sure if they can be unified.

Quoted 4 lines

Regarding Andy's comment, I think he's referring to the duplicated chip infos.
In the case of the bmp280, the chip_infos are defined on the common driver code and used for both SPI and I2C match tables.

Hi,

I'm loosing track of where we are with this driver as multiple people are working on it.

Quoted 6 lines
Quoted 3 lines

Angel, as most of the work is yours, please could you manage the flow of patches for this one so I get series with clear statement of what they are dependent on.

Sure. If Biju is okay with it, maybe I should squash toghether this two series of mine:

I am ok with it, as I don't have bandwidth as well as board for testing it. Please feel free to post.

Cheers,
Biju

#7 lore raw

Re: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

In reply to Jonathan Cameron
Quoted 65 lines · Jonathan Cameron
Quoted 55 lines · On Mon, 14 Aug 2023 18:43:49 +0200 Angel Iglesias <[email protected]>…
Quoted 38 lines · Biju Das

Hi Angel Iglesias,

Quoted 31 lines

Subject: Re: [PATCH v2 2/3] iio: pressure: bmp280: Use i2c_get_match_data

Quoted 21 lines · Angel Iglesias

Replaces device_get_match_data() and fallback match_id logic by new unified helper function i2c_get_match_data().

(quoted diff) +0 -0

b/drivers/iio/pressure/bmp280- i2c.c index 693eb1975fdc..34e3bc758493 100644

drivers/iio/pressure/bmp280-i2c.c +1 -3
@@ -11,9 +11,7 @@ static int bmp280_i2c_probe(struct i2c_client
 *client)
         const struct bmp280_chip_info *chip_info;
         struct regmap *regmap;
 
-       chip_info = device_get_match_data(&client->dev);
-       if (!chip_info)
-               chip_info = (const struct bmp280_chip_info *)
 id->driver_data;
+       chip_info = i2c_get_match_data(client);
 
Quoted 3 lines

        regmap = devm_regmap_init_i2c(client,
chip_info->regmap_config);
        if (IS_ERR(regmap)) {

Hi,

Quoted 3 lines

I noticed I submitted this change that was also submitted by Biju Das on another
patch:

Quoted 8 lines
Quoted 1 line

Should I drop this patch from the series?

I think it is ok. Andy is suggesting to use unified table for SPI/I2C

Is it something do able and testable in your environment? see [1], If yes, please create another patch for using unified table for both i2c and
spi.

I have around a BMP390 with the SPI pins available to test it out. In the case
of the bmp280 we could unify the of_match table as they're almost the same. In
the case of the spi_device_id and i2c_device_id tables, as they're different structs I'm not sure if they can be unified.

Quoted 5 lines

Regarding Andy's comment, I think he's referring to the duplicated chip infos.
In the case of the bmp280, the chip_infos are defined on the common driver code
and used for both SPI and I2C match tables.

Hi,

Quoted 7 lines

I'm loosing track of where we are with this driver as multiple people are working on it.

Angel, as most of the work is yours, please could you manage the flow of patches
for this one so I get series with clear statement of what they are dependent on.

Sure. If Biju is okay with it, maybe I should squash toghether this two series of mine:
https://patchwork.kernel.org/project/linux-iio/cover/[email protected]/
https://patchwork.kernel.org/project/linux-iio/cover/[email protected]/

And pull this patch from Biju:
https://patchwork.kernel.org/project/linux-iio/patch/[email protected]/

Sorry again from the noise introduced in the mail-list,

Kind regards,
Angel

Quoted 15 lines

Thanks,

Jonathan

Quoted 10 lines
Quoted 6 lines

https://lore.kernel.org/linux-renesas-soc/CAHp75VeX+T=hAN+PgtHTdv4b6UtDVgveUUww1b1kuOngzDinFw@mail.gmail.com/T/#t

Cheers,
Biju

Kind regards,
Angel