ti-ads7950: fix gpio handling and facelift
v2 cover ti-ads7950: fix gpio handling and facelift
The original reason for this series is to make sure ti_ads7950_get() returns result in range [0, 1] or a negative error code to comply with gpiolib requirement.
During review David noticed the fact that the function also clobbers gpio state. Fixing that lead to adding cleanups using guard() and devm.
v2:
- style fixes in #1 (per David)
- fix to not clobber state (#2, new)
- #3 and #4 (new)
Dmitry Torokhov (4):
iio: adc: ti-ads7950: normalize return value of gpio_get
iio: adc: ti-ads7950: do not clobber gpio state in ti_ads7950_get()
iio: adc: ti-ads7950: switch to using guard() notation
iio: adc: ti-ads7950: complete conversion to using managed resources
drivers/iio/adc/ti-ads7950.c | 206 +++++++++++++++-------------------- 1 file changed, 90 insertions(+), 116 deletions(-)
Thanks.
Signature · 1 line
Dmitry
PATCH v2 1/4 iio: adc: ti-ads7950: normalize return value of gpio_get
The GPIO get callback is expected to return 0 or 1 (or a negative error code). Ensure that the value returned by ti_ads7950_get() for output pins is normalized to the [0, 1] range.
--- drivers/iio/adc/ti-ads7950.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/adc/ti-ads7950.c +1 -1
@@ -433,7 +433,7 @@ static int ti_ads7950_get(struct gpio_chip *chip, unsigned int offset)/* If set as output, return the output */ if (st->gpio_cmd_settings_bitmask &BIT (offset)) { - ret = st->cmd_settings_bitmask &BIT (offset); + ret = (st->cmd_settings_bitmask &BIT (offset)) ?1 :0 ; goto out; }
Signature · 1 line
2.53.0.335.g19a08e0c02-goog
Re: [PATCH v2 1/4] iio: adc: ti-ads7950: normalize return value of gpio_get
Quoted 3 lines · Dmitry Torokhov
The GPIO get callback is expected to return 0 or 1 (or a negative error code). Ensure that the value returned by
ti_ads7950_get()for output pins is normalized to the [0, 1] range.
Signature · 2 lines
With Best Regards, Andy Shevchenko
Re: [PATCH v2 1/4] iio: adc: ti-ads7950: normalize return value of gpio_get
In reply to Dmitry TorokhovQuoted 23 lines · Dmitry Torokhov
The GPIO get callback is expected to return 0 or 1 (or a negative error code). Ensure that the value returned by
ti_ads7950_get()for output pins is normalized to the [0, 1] range.Signed-off-by Dmitry Torokhov <[email protected]>--- drivers/iio/adc/ti-ads7950.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/iio/adc/ti-ads7950.c +1
-2@@ -433,7 +433,7 @@ static int ti_ads7950_get(struct gpio_chip *chip, unsigned int offset)/* If set as output, return the output */ if (st->gpio_cmd_settings_bitmask &BIT (offset)) { - ret = st->cmd_settings_bitmask &BIT (offset); + ret = (st->cmd_settings_bitmask &BIT (offset)) ?1 :0 ; goto out; } --
Quoted 3 lines
2.53.0.335.g19a08e0c02-goog
Re: [PATCH v2 1/4] iio: adc: ti-ads7950: normalize return value of gpio_get
In reply to Dmitry TorokhovQuoted 5 lines · Dmitry Torokhov
The GPIO get callback is expected to return 0 or 1 (or a negative error code). Ensure that the value returned by
ti_ads7950_get()for output pins is normalized to the [0, 1] range.Signed-off-by Dmitry Torokhov <[email protected]>
Yours,
Linus Walleij
Jonathan Cameron
maintainerRe: [PATCH v2 1/4] iio: adc: ti-ads7950: normalize return value of gpio_get
In reply to Dmitry TorokhovQuoted 3 lines · On Wed, 18 Feb 2026 18:29:25 -0800 Dmitry Torokhov <[email protected]>…
The GPIO get callback is expected to return 0 or 1 (or a negative error code). Ensure that the value returned by
ti_ads7950_get()for output pins is normalized to the [0, 1] range.
Fixes? I'm not quite sure with something that says 'expected'
Jonathan
Quoted 19 lines
Signed-off-by Dmitry Torokhov <[email protected]>--- drivers/iio/adc/ti-ads7950.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/iio/adc/ti-ads7950.c +1
-1@@ -433,7 +433,7 @@ static int ti_ads7950_get(struct gpio_chip *chip, unsigned int offset)/* If set as output, return the output */ if (st->gpio_cmd_settings_bitmask &BIT (offset)) { - ret = st->cmd_settings_bitmask &BIT (offset); + ret = (st->cmd_settings_bitmask &BIT (offset)) ?1 :0 ; goto out; }
Re: [PATCH v2 1/4] iio: adc: ti-ads7950: normalize return value of gpio_get
Quoted 7 lines · Jonathan Cameron
Quoted 3 lines · On Wed, 18 Feb 2026 18:29:25 -0800 Dmitry Torokhov <[email protected]>…
The GPIO get callback is expected to return 0 or 1 (or a negative error code). Ensure that the value returned by
ti_ads7950_get()for output pins is normalized to the [0, 1] range.Fixes? I'm not quite sure with something that says 'expected'
Will add a tag.
Thanks.
Signature · 1 line
Dmitry
PATCH v2 2/4 iio: adc: ti-ads7950: do not clobber gpio state in ti_ads7950_get()
In reply to Dmitry TorokhovGPIO state was inadvertently overwritten by the result of sip_sync, reuniting in ti_ads7950_get() only returning 0 as gpio state (or error).
Fix this by introducing a separate variable to hold the state.
--- drivers/iio/adc/ti-ads7950.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
drivers/iio/adc/ti-ads7950.c +5 -4
@@ -427,13 +427,14 @@ static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset,static int ti_ads7950_get (struct gpio_chip *chip,unsigned int offset) {struct ti_ads7950_state *st =gpiochip_get_data (chip); -int ret; +int ret =0 ; +bool state;mutex_lock (&st->slock);/* If set as output, return the output */ if (st->gpio_cmd_settings_bitmask &BIT (offset)) { - ret = (st->cmd_settings_bitmask &BIT (offset)) ?1 :0 ; + state = st->cmd_settings_bitmask &BIT (offset); goto out; }
@@ -444,7 +445,7 @@ static int ti_ads7950_get(struct gpio_chip *chip, unsigned int offset)if (ret) goto out; - ret = ((st->single_rx >>12 ) &BIT (offset)) ?1 :0 ; + state = (st->single_rx >>12 ) &BIT (offset);/* Revert back to original settings */ st->cmd_settings_bitmask &= ~TI_ADS7950_CR_GPIO_DATA ;
@@ -456,7 +457,7 @@ static int ti_ads7950_get(struct gpio_chip *chip, unsigned int offset) out:mutex_unlock (&st->slock); -return ret; +return ret ?: state; } static int ti_ads7950_get_direction(struct gpio_chip *chip,
Signature · 1 line
2.53.0.335.g19a08e0c02-goog
Re: [PATCH v2 2/4] iio: adc: ti-ads7950: do not clobber gpio state in ti_ads7950_get()
Quoted 1 line · Dmitry Torokhov
GPIO state was inadvertently overwritten by the result of
sip_sync,
sip?
Quoted 1 line
reuniting in
ti_ads7950_get()only returning 0 as gpio state (or error).
GPIO
Quoted 1 line
Fix this by introducing a separate variable to hold the state.
...
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset, · line 430
Quoted 2 lines
drivers/iio/adc/ti-ads7950.c +1
-1-int ret; +int ret =0 ;
This kind of assignment is harder to maintain, because it leaves a room for subtle mistakes (when it's reused by a newly injected code, quite similar, actually, to what this change tries to address). Please, decouple definition and assignment.
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset, · line 431
Quoted 1 line
drivers/iio/adc/ti-ads7950.c +1
-0+bool state;
Not sure about this (yes, I know it was suggested). I would leave it still the same type as one that is returned by the function or a compatible one that is the same as st->* (if the used ones are all of the same type).
Signature · 2 lines
With Best Regards, Andy Shevchenko
David Lechner
reviewerRe: [PATCH v2 2/4] iio: adc: ti-ads7950: do not clobber gpio state in ti_ads7950_get()
In reply to Dmitry TorokhovQuoted 6 lines · Dmitry Torokhov
GPIO state was inadvertently overwritten by the result of
sip_sync, reuniting inti_ads7950_get()only returning 0 as gpio state (or error).Fix this by introducing a separate variable to hold the state.
Reported-by David Lechner <[email protected]>
This should have a Fixes: tag since it is fixing a real bug.
Also, fixes should come first in the series.
Quoted 13 lines
Signed-off-by Dmitry Torokhov <[email protected]>--- drivers/iio/adc/ti-ads7950.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)drivers/iio/adc/ti-ads7950.c +0
-0@@ -427,13 +427,14 @@ static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset,static int ti_ads7950_get (struct gpio_chip *chip,unsigned int offset) {struct ti_ads7950_state *st =gpiochip_get_data (chip);
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset, · line 437
Quoted 10 lines
drivers/iio/adc/ti-ads7950.c +3
-2-int ret; +int ret =0 ; +bool state;mutex_lock (&st->slock);/* If set as output, return the output */ if (st->gpio_cmd_settings_bitmask &BIT (offset)) { - ret = (st->cmd_settings_bitmask &BIT (offset)) ?1 :0 ; + state = st->cmd_settings_bitmask &BIT (offset);
I agree it would be better to put...
ret = 0;
here.
Quoted 20 lines
drivers/iio/adc/ti-ads7950.c +2
-2goto out; }@@ -444,7 +445,7 @@ static int ti_ads7950_get(struct gpio_chip *chip, unsigned int offset)if (ret) goto out; - ret = ((st->single_rx >>12 ) &BIT (offset)) ?1 :0 ; + state = (st->single_rx >>12 ) &BIT (offset);/* Revert back to original settings */ st->cmd_settings_bitmask &= ~TI_ADS7950_CR_GPIO_DATA ;@@ -456,7 +457,7 @@ static int ti_ads7950_get(struct gpio_chip *chip, unsigned int offset) out:mutex_unlock (&st->slock); -return ret; + return ret ?: state;> } static int ti_ads7950_get_direction(struct gpio_chip *chip,
PATCH v2 3/4 iio: adc: ti-ads7950: switch to using guard() notation
In reply to Dmitry Torokhovguard() notation allows early returns when encountering errors, making control flow more obvious. Use it.
Also variables that now only hold error codes (or 0) are renamed to "error" to make their purpose clearer.
--- drivers/iio/adc/ti-ads7950.c | 105 ++++++++++++++++------------------- 1 file changed, 48 insertions(+), 57 deletions(-)
drivers/iio/adc/ti-ads7950.c +48 -57
@@ -306,18 +306,17 @@ static irqreturn_t ti_ads7950_trigger_handler(int irq, void *p)struct iio_poll_func *pf = p;struct iio_dev *indio_dev = pf->indio_dev;struct ti_ads7950_state *st =iio_priv (indio_dev); -int ret; +int error; -mutex_lock (&st->slock); - ret =spi_sync (st->spi, &st->ring_msg); -if (ret <0 ) - goto out; + scoped_guard(mutex, &st->slock) { + error =spi_sync (st->spi, &st->ring_msg); +if (error) +break ; -iio_push_to_buffers_with_timestamp (indio_dev, &st->rx_buf[2 ], -iio_get_time_ns (indio_dev)); +iio_push_to_buffers_with_timestamp (indio_dev, &st->rx_buf[2 ], +iio_get_time_ns (indio_dev)); + } -out: -mutex_unlock (&st->slock);iio_trigger_notify_done (indio_dev->trig);return IRQ_HANDLED ;
@@ -326,22 +325,19 @@ static irqreturn_t ti_ads7950_trigger_handler(int irq, void *p)static int ti_ads7950_scan_direct (struct iio_dev *indio_dev,unsigned int ch) {struct ti_ads7950_state *st =iio_priv (indio_dev); -int ret, cmd; +int error; +int cmd; + +guard (mutex)(&st->slock); -mutex_lock (&st->slock); cmd =TI_ADS7950_MAN_CMD (TI_ADS7950_CR_CHAN (ch)); st->single_tx = cmd; - ret =spi_sync (st->spi, &st->scan_single_msg); -if (ret) - goto out; - - ret = st->single_rx; - -out: -mutex_unlock (&st->slock); + error =spi_sync (st->spi, &st->scan_single_msg); +if (error) +return error; -return ret; +return st->single_rx; }static int ti_ads7950_get_range (struct ti_ads7950_state *st)
@@ -407,9 +403,9 @@ static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset, int value) {struct ti_ads7950_state *st =gpiochip_get_data (chip); -int ret; +int error; -mutex_lock (&st->slock); +guard (mutex)(&st->slock);if (value) st->cmd_settings_bitmask |=BIT (offset);
@@ -417,47 +413,44 @@ static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset, st->cmd_settings_bitmask &= ~BIT (offset); st->single_tx =TI_ADS7950_MAN_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); - -mutex_unlock (&st->slock); + error =spi_sync (st->spi, &st->scan_single_msg); +if (error) +return error; -return ret; +return 0 ; }static int ti_ads7950_get (struct gpio_chip *chip,unsigned int offset) {struct ti_ads7950_state *st =gpiochip_get_data (chip); -int ret =0 ;bool state; +int error; -mutex_lock (&st->slock); +guard (mutex)(&st->slock);/* If set as output, return the output */ if (st->gpio_cmd_settings_bitmask &BIT (offset)) { state = st->cmd_settings_bitmask &BIT (offset); - goto out; +return state; }/* GPIO data bit sets SDO bits 12-15 to GPIO input */ st->cmd_settings_bitmask |=TI_ADS7950_CR_GPIO_DATA ; st->single_tx =TI_ADS7950_MAN_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); -if (ret) - goto out; + error =spi_sync (st->spi, &st->scan_single_msg); +if (error) +return error; state = (st->single_rx >>12 ) &BIT (offset);/* Revert back to original settings */ st->cmd_settings_bitmask &= ~TI_ADS7950_CR_GPIO_DATA ; st->single_tx =TI_ADS7950_MAN_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); -if (ret) - goto out; - -out: -mutex_unlock (&st->slock); + error =spi_sync (st->spi, &st->scan_single_msg); +if (error) +return error; -return ret ?: state; +return state; } static int ti_ads7950_get_direction(struct gpio_chip *chip,
@@ -473,9 +466,9 @@ static int _ti_ads7950_set_direction(struct gpio_chip *chip, int offset, int input) {struct ti_ads7950_state *st =gpiochip_get_data (chip); -int ret =0 ; +int error; -mutex_lock (&st->slock); +guard (mutex)(&st->slock);/* Only change direction if needed */ if (input && (st->gpio_cmd_settings_bitmask &BIT (offset)))
@@ -483,15 +476,14 @@ static int _ti_ads7950_set_direction(struct gpio_chip *chip, int offset,else if (!input && !(st->gpio_cmd_settings_bitmask &BIT (offset))) st->gpio_cmd_settings_bitmask |=BIT (offset);else - goto out; +return 0 ; st->single_tx =TI_ADS7950_GPIO_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); + error =spi_sync (st->spi, &st->scan_single_msg); +if (error) +return error; -out: -mutex_unlock (&st->slock); - -return ret; +return 0 ; } static int ti_ads7950_direction_input(struct gpio_chip *chip,
@@ -514,27 +506,26 @@ static int ti_ads7950_direction_output(struct gpio_chip *chip,static int ti_ads7950_init_hw (struct ti_ads7950_state *st) { -int ret =0 ; +int error; -mutex_lock (&st->slock); +guard (mutex)(&st->slock);/* Settings for Manual/Auto1/Auto2 commands */ /* Default to 5v ref */ st->cmd_settings_bitmask =TI_ADS7950_CR_RANGE_5V ; st->single_tx =TI_ADS7950_MAN_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); -if (ret) - goto out; + error =spi_sync (st->spi, &st->scan_single_msg); +if (error) +return error;/* Settings for GPIO command */ st->gpio_cmd_settings_bitmask =0x0 ; st->single_tx =TI_ADS7950_GPIO_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); - -out: -mutex_unlock (&st->slock); + error =spi_sync (st->spi, &st->scan_single_msg); +if (error) +return error; -return ret; +return 0 ; }static int ti_ads7950_probe (struct spi_device *spi)
Signature · 1 line
2.53.0.335.g19a08e0c02-goog
Re: [PATCH v2 3/4] iio: adc: ti-ads7950: switch to using guard() notation
Quoted 1 line · Dmitry Torokhov
guard()notation allows early returns when encountering errors, making
guard()()
// strictly speaking
Quoted 1 line
control flow more obvious. Use it.
I like the change, but...
Quoted 2 lines
Also variables that now only hold error codes (or 0) are renamed to "error" to make their purpose clearer.
...this does not belong to the patch. If you wish to rename, better doing it separately. This, in particular, adds undesired churn in the change making it unclear.
Signature · 2 lines
With Best Regards, Andy Shevchenko
David Lechner
reviewerRe: [PATCH v2 3/4] iio: adc: ti-ads7950: switch to using guard() notation
Quoted 13 lines · Andy Shevchenko
Quoted 1 line · Dmitry Torokhov
guard()notation allows early returns when encountering errors, makingguard()()
// strictly speaking
Quoted 1 line
control flow more obvious. Use it.
I like the change, but...
Quoted 2 lines
Also variables that now only hold error codes (or 0) are renamed to "error" to make their purpose clearer.
Normally I would not give my opinion on this, but since I wrote the driver originally, I will say please don't rename. I prefer to always use "ret".
Quoted 5 lines
...this does not belong to the patch. If you wish to rename, better doing it separately. This, in particular, adds undesired churn in the change making it unclear.
Re: [PATCH v2 3/4] iio: adc: ti-ads7950: switch to using guard() notation
Quoted 10 lines · David Lechner
Quoted 9 lines · Andy Shevchenko
Quoted 1 line · Dmitry Torokhov
guard()notation allows early returns when encountering errors, makingguard()()
// strictly speaking
Quoted 1 line
control flow more obvious. Use it.
Quoted 7 lines
Quoted 4 lines
I like the change, but...
Quoted 2 lines
Also variables that now only hold error codes (or 0) are renamed to "error" to make their purpose clearer.
Normally I would not give my opinion on this, but since I wrote the driver originally, I will say please don't rename. I prefer to always use "ret".
I hope I can convince you otherwise.
IMO "ret" or "retval" should be used when the returned value is intended to be used during normal operation. For cases where we only expect to have an error or 0 for success "error" or "err" is more appropriate. This allows you to write
error = do_action(...);
if (error) {
// handle error somehow, typically simply report.
}This also helps when reading the code as you know that there is usually no reason to care about the specific value in this variable (maybe except -EPROBE_DEFER).
I will push the conversion ret -> error to the very last patch so it can easily be dropped if I am unsuccessful in swaying your opinion.
Thanks.
Signature · 1 line
Dmitry
David Lechner
reviewerRe: [PATCH v2 3/4] iio: adc: ti-ads7950: switch to using guard() notation
Quoted 13 lines · Dmitry Torokhov
Quoted 12 lines · David Lechner
Quoted 11 lines · Andy Shevchenko
Quoted 1 line · Dmitry Torokhov
guard()notation allows early returns when encountering errors, makingguard()()
// strictly speaking
Quoted 1 line
control flow more obvious. Use it.
I like the change, but...
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines
Quoted 2 lines
Also variables that now only hold error codes (or 0) are renamed to "error" to make their purpose clearer.
Normally I would not give my opinion on this, but since I wrote the driver originally, I will say please don't rename. I prefer to always use "ret".
I hope I can convince you otherwise.
I'm afraid not. "ret" is used > 35k times in IIO and err is used < 3k times and error < 1k times. So I am really used to seeing only "ret" for errors and "error" looks very unusual to my eyes because of this.
You mentioned valuing the common pattern in your other reply, so I hope you can understand that this is what I value more here.
Quoted 11 lines
IMO "ret" or "retval" should be used when the returned value is intended to be used during normal operation. For cases where we only expect to have an error or 0 for success "error" or "err" is more appropriate. This allows you to write
error = do_action(...); if (error) { // handle error somehow, typically simply report. }
Quoted 9 lines
This also helps when reading the code as you know that there is usually no reason to care about the specific value in this variable (maybe except -EPROBE_DEFER).
I will push the conversion ret -> error to the very last patch so it can easily be dropped if I am unsuccessful in swaying your opinion.
Thanks.
To give you fair warning, it will still be NAK from me.
David Lechner
reviewerRe: [PATCH v2 3/4] iio: adc: ti-ads7950: switch to using guard() notation
In reply to Dmitry TorokhovQuoted 20 lines · Dmitry Torokhov
guard()notation allows early returns when encountering errors, making control flow more obvious. Use it.Also variables that now only hold error codes (or 0) are renamed to "error" to make their purpose clearer.
Signed-off-by Dmitry Torokhov <[email protected]>--- drivers/iio/adc/ti-ads7950.c | 105 ++++++++++++++++------------------- 1 file changed, 48 insertions(+), 57 deletions(-)drivers/iio/adc/ti-ads7950.c +0
-1@@ -306,18 +306,17 @@ static irqreturn_t ti_ads7950_trigger_handler(int irq, void *p)struct iio_poll_func *pf = p;struct iio_dev *indio_dev = pf->indio_dev;struct ti_ads7950_state *st =iio_priv (indio_dev); -int ret;
drivers/iio/adc/ti-ads7950.c static irqreturn_t ti_ads7950_trigger_handler(int irq, void *p) · line 314
Quoted 10 lines
drivers/iio/adc/ti-ads7950.c +5
-4+int error; -mutex_lock (&st->slock); - ret =spi_sync (st->spi, &st->ring_msg); -if (ret <0 ) - goto out; + scoped_guard(mutex, &st->slock) { + error =spi_sync (st->spi, &st->ring_msg); +if (error) +break ;
I'm not a fan of scoped_guard() because of the hidden for loop in it. It hides the fact that the break; is breaking out of that for loop.
It would be more clear/obvious written as:
do {
guard(mutex)(&st->slock);
ret = spi_sync(st->spi, &st->ring_msg);
if (ret)
break;
iio_push_to_buffers_with_timestamp(indio_dev, &st->rx_buf[2],
iio_get_time_ns(indio_dev));
} while (0);Quoted 75 lines
drivers/iio/adc/ti-ads7950.c +17
-22-iio_push_to_buffers_with_timestamp (indio_dev, &st->rx_buf[2 ], -iio_get_time_ns (indio_dev)); +iio_push_to_buffers_with_timestamp (indio_dev, &st->rx_buf[2 ], +iio_get_time_ns (indio_dev)); + } -out: -mutex_unlock (&st->slock);iio_trigger_notify_done (indio_dev->trig);return IRQ_HANDLED ;@@ -326,22 +325,19 @@ static irqreturn_t ti_ads7950_trigger_handler(int irq, void *p)static int ti_ads7950_scan_direct (struct iio_dev *indio_dev,unsigned int ch) {struct ti_ads7950_state *st =iio_priv (indio_dev); -int ret, cmd; +int error; +int cmd; + +guard (mutex)(&st->slock); -mutex_lock (&st->slock); cmd =TI_ADS7950_MAN_CMD (TI_ADS7950_CR_CHAN (ch)); st->single_tx = cmd; - ret =spi_sync (st->spi, &st->scan_single_msg); -if (ret) - goto out; - - ret = st->single_rx; - -out: -mutex_unlock (&st->slock); + error =spi_sync (st->spi, &st->scan_single_msg); +if (error) +return error; -return ret; +return st->single_rx; }static int ti_ads7950_get_range (struct ti_ads7950_state *st)@@ -407,9 +403,9 @@ static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset, int value) {struct ti_ads7950_state *st =gpiochip_get_data (chip); -int ret; +int error; -mutex_lock (&st->slock); +guard (mutex)(&st->slock);if (value) st->cmd_settings_bitmask |=BIT (offset);@@ -417,47 +413,44 @@ static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset, st->cmd_settings_bitmask &= ~BIT (offset); st->single_tx =TI_ADS7950_MAN_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); - -mutex_unlock (&st->slock); + error =spi_sync (st->spi, &st->scan_single_msg); +if (error) +return error; -return ret; +return 0 ; }static int ti_ads7950_get (struct gpio_chip *chip,unsigned int offset) {struct ti_ads7950_state *st =gpiochip_get_data (chip); -int ret =0 ;bool state;
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset, · line 434
Quoted 10 lines
drivers/iio/adc/ti-ads7950.c +3
-2+int error; -mutex_lock (&st->slock); +guard (mutex)(&st->slock);/* If set as output, return the output */ if (st->gpio_cmd_settings_bitmask &BIT (offset)) { state = st->cmd_settings_bitmask &BIT (offset); - goto out; +return state;
This can return directly instead of using local variable.
Quoted 45 lines
drivers/iio/adc/ti-ads7950.c +9
-12}/* GPIO data bit sets SDO bits 12-15 to GPIO input */ st->cmd_settings_bitmask |=TI_ADS7950_CR_GPIO_DATA ; st->single_tx =TI_ADS7950_MAN_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); -if (ret) - goto out; + error =spi_sync (st->spi, &st->scan_single_msg); +if (error) +return error; state = (st->single_rx >>12 ) &BIT (offset);/* Revert back to original settings */ st->cmd_settings_bitmask &= ~TI_ADS7950_CR_GPIO_DATA ; st->single_tx =TI_ADS7950_MAN_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); -if (ret) - goto out; - -out: -mutex_unlock (&st->slock); + error =spi_sync (st->spi, &st->scan_single_msg); +if (error) +return error; -return ret ?: state; +return state; } static int ti_ads7950_get_direction(struct gpio_chip *chip,@@ -473,9 +466,9 @@ static int _ti_ads7950_set_direction(struct gpio_chip *chip, int offset, int input) {struct ti_ads7950_state *st =gpiochip_get_data (chip); -int ret =0 ; +int error; -mutex_lock (&st->slock); +guard (mutex)(&st->slock);/* Only change direction if needed */ if (input && (st->gpio_cmd_settings_bitmask &BIT (offset)))
drivers/iio/adc/ti-ads7950.c static int _ti_ads7950_set_direction(struct gpio_chip *chip, int offset, · line 482
Quoted 9 lines
drivers/iio/adc/ti-ads7950.c +2
-2@@ -483,15 +476,14 @@ static int _ti_ads7950_set_direction(struct gpio_chip *chip, int offset,else if (!input && !(st->gpio_cmd_settings_bitmask &BIT (offset))) st->gpio_cmd_settings_bitmask |=BIT (offset);else - goto out; +return 0 ; st->single_tx =TI_ADS7950_GPIO_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); + error =spi_sync (st->spi, &st->scan_single_msg);
Can just return directly here now.
Quoted 30 lines
drivers/iio/adc/ti-ads7950.c +7
-9+if (error) +return error; -out: -mutex_unlock (&st->slock); - -return ret; +return 0 ; } static int ti_ads7950_direction_input(struct gpio_chip *chip,@@ -514,27 +506,26 @@ static int ti_ads7950_direction_output(struct gpio_chip *chip,static int ti_ads7950_init_hw (struct ti_ads7950_state *st) { -int ret =0 ; +int error; -mutex_lock (&st->slock); +guard (mutex)(&st->slock);/* Settings for Manual/Auto1/Auto2 commands */ /* Default to 5v ref */ st->cmd_settings_bitmask =TI_ADS7950_CR_RANGE_5V ; st->single_tx =TI_ADS7950_MAN_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); -if (ret) - goto out; + error =spi_sync (st->spi, &st->scan_single_msg); +if (error)
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_direction_output(struct gpio_chip *chip, · line 524
Quoted 10 lines
drivers/iio/adc/ti-ads7950.c +2
-4+return error;/* Settings for GPIO command */ st->gpio_cmd_settings_bitmask =0x0 ; st->single_tx =TI_ADS7950_GPIO_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); - -out: -mutex_unlock (&st->slock); + error =spi_sync (st->spi, &st->scan_single_msg);
And can return directly here.
Quoted 8 lines
drivers/iio/adc/ti-ads7950.c +3
-1+if (error) +return error; -return ret; +return 0 ; }static int ti_ads7950_probe (struct spi_device *spi)
Re: [PATCH v2 3/4] iio: adc: ti-ads7950: switch to using guard() notation
Quoted 11 lines · David Lechner
drivers/iio/adc/ti-ads7950.c static irqreturn_t ti_ads7950_trigger_handler(int irq, void *p) · line 314
Quoted 4 lines · Dmitry Torokhov
drivers/iio/adc/ti-ads7950.c +4
-0+ scoped_guard(mutex, &st->slock) { + error =spi_sync (st->spi, &st->ring_msg); +if (error) +break ;I'm not a fan of
scoped_guard()because of the hidden for loop in it. It hides the fact that the break; is breaking out of that for loop.It would be more clear/obvious written as:
Quoted 10 lines
do {
guard(mutex)(&st->slock);
ret = spi_sync(st->spi, &st->ring_msg);
if (ret)
break;
iio_push_to_buffers_with_timestamp(indio_dev, &st->rx_buf[2],
iio_get_time_ns(indio_dev));
} while (0);OK.
I could also make it
scoped_guard(mutex, &st->slock) {
ret = spi_sync(st->spi, &st->ring_msg);
if (!ret)
iio_push_to_buffers_with_timestamp(indio_dev, &st->rx_buf[2],
iio_get_time_ns(indio_dev));
}to avoid using "break".
I think you will find that scoped_guard() will gain the foothold in the kernel so having implementation that does not follow common pattern might not be the best option.
Quoted 8 lines
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset, · line 434
Quoted 6 lines
drivers/iio/adc/ti-ads7950.c +1
-1/* If set as output, return the output */ if (st->gpio_cmd_settings_bitmask &BIT (offset)) { state = st->cmd_settings_bitmask &BIT (offset); - goto out; +return state;This can return directly instead of using local variable.
This will require the explicitly normalizing, which we avoided by introducing "bool state" to begin with...
Quoted 6 lines
drivers/iio/adc/ti-ads7950.c static int _ti_ads7950_set_direction(struct gpio_chip *chip, int offset, · line 482
Quoted 4 lines
drivers/iio/adc/ti-ads7950.c +1
-1st->single_tx =TI_ADS7950_GPIO_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); + error =spi_sync (st->spi, &st->scan_single_msg);Can just return directly here now.
I think there is benefit in explicitly calling out the error paths and explicitly return 0 on success. It removes the doubt whether a function can return positive value on success.
Thanks.
Signature · 1 line
Dmitry
David Lechner
reviewerRe: [PATCH v2 3/4] iio: adc: ti-ads7950: switch to using guard() notation
Quoted 25 lines · Dmitry Torokhov
Quoted 11 lines · David Lechner
drivers/iio/adc/ti-ads7950.c static irqreturn_t ti_ads7950_trigger_handler(int irq, void *p) · line 314
Quoted 4 lines · Dmitry Torokhov
drivers/iio/adc/ti-ads7950.c +4
-0+ scoped_guard(mutex, &st->slock) { + error =spi_sync (st->spi, &st->ring_msg); +if (error) +break ;I'm not a fan of
scoped_guard()because of the hidden for loop in it. It hides the fact that the break; is breaking out of that for loop.It would be more clear/obvious written as:
Quoted 10 lines
do { guard(mutex)(&st->slock); ret = spi_sync(st->spi, &st->ring_msg); if (ret) break; iio_push_to_buffers_with_timestamp(indio_dev, &st->rx_buf[2], iio_get_time_ns(indio_dev)); } while (0);OK.
Quoted 8 lines
I could also make it
scoped_guard(mutex, &st->slock) { ret = spi_sync(st->spi, &st->ring_msg); if (!ret) iio_push_to_buffers_with_timestamp(indio_dev, &st->rx_buf[2], iio_get_time_ns(indio_dev)); }
This is OK.
Quoted 6 lines
to avoid using "break".
I think you will find that
scoped_guard()will gain the foothold in the kernel so having implementation that does not follow common pattern might not be the best option.
In general, I agree with this idea.
In this case, the common pattern unfortunately leads to common bugs. Since we were discussing this, I audited the kernel and found and reported 3 unintentional mistakes with break/continue when scoped_guard() was introduced. And I have caught more in reviews before they made it into the kernel.
So I make an exception here to thinking that the common pattern is best.
Quoted 20 lines
Quoted 9 lines
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_set(struct gpio_chip *chip, unsigned int offset, · line 434
Quoted 7 lines
drivers/iio/adc/ti-ads7950.c +1
-1/* If set as output, return the output */ if (st->gpio_cmd_settings_bitmask &BIT (offset)) { state = st->cmd_settings_bitmask &BIT (offset); - goto out; +return state;This can return directly instead of using local variable.
This will require the explicitly normalizing, which we avoided by introducing "bool state" to begin with...
Quoted 6 lines
drivers/iio/adc/ti-ads7950.c static int _ti_ads7950_set_direction(struct gpio_chip *chip, int offset, · line 482
Quoted 4 lines
drivers/iio/adc/ti-ads7950.c +1
-1st->single_tx =TI_ADS7950_GPIO_CMD_SETTINGS (st); - ret =spi_sync (st->spi, &st->scan_single_msg); + error =spi_sync (st->spi, &st->scan_single_msg);Can just return directly here now.
Quoted 3 lines
I think there is benefit in explicitly calling out the error paths and explicitly return 0 on success. It removes the doubt whether a function can return positive value on success.
In the IIO subsystem, the direct return is preferred (maintainer and other reviewers frequently request this).
Quoted 3 lines
Thanks.
PATCH v2 4/4 iio: adc: ti-ads7950: complete conversion to using managed resources
In reply to Dmitry TorokhovAll resources that the driver needs have managed API now. Switch to using them to make code clearer and drop ti_ads7950_remove().
--- drivers/iio/adc/ti-ads7950.c | 98 +++++++++++++++--------------------- 1 file changed, 40 insertions(+), 58 deletions(-)
drivers/iio/adc/ti-ads7950.c +40 -58
@@ -528,19 +528,26 @@ static int ti_ads7950_init_hw(struct ti_ads7950_state *st)return 0 ; } +static void ti_ads7950_power_off (void *data) +{ +struct ti_ads7950_state *st = data; + +regulator_disable (st->reg); +} +static int ti_ads7950_probe (struct spi_device *spi) {struct ti_ads7950_state *st;struct iio_dev *indio_dev;const struct ti_ads7950_chip_info *info; -int ret; +int error; spi->bits_per_word =16 ; spi->mode |=SPI_CS_WORD ; - ret =spi_setup (spi); -if (ret <0 ) { + error =spi_setup (spi); +if (error) {dev_err (&spi->dev,"Error in spi setup\n" ); -return ret; +return error; } indio_dev =devm_iio_device_alloc (&spi->dev,sizeof (*st));
@@ -598,36 +605,36 @@ static int ti_ads7950_probe(struct spi_device *spi)mutex_init (&st->slock); st->reg =devm_regulator_get (&spi->dev,"vref" ); -if (IS_ERR (st->reg)) { - ret =dev_err_probe (&spi->dev,PTR_ERR (st->reg), + error =PTR_ERR_OR_ZERO (st->reg); +if (error) +return dev_err_probe (&spi->dev, error,"Failed to get regulator \"vref\"\n" ); - goto error_destroy_mutex; - } - ret =regulator_enable (st->reg); -if (ret) { -dev_err (&spi->dev,"Failed to enable regulator \"vref\"\n" ); - goto error_destroy_mutex; - } + error =regulator_enable (st->reg); +if (error) +return dev_err_probe (&spi->dev, error, +"Failed to enable regulator \"vref\"\n" ); - ret =iio_triggered_buffer_setup (indio_dev,NULL , - &ti_ads7950_trigger_handler,NULL ); -if (ret) { -dev_err (&spi->dev,"Failed to setup triggered buffer\n" ); - goto error_disable_reg; - } + error =devm_add_action_or_reset (&spi->dev, ti_ads7950_power_off, st); +if (error) +return error; - ret =ti_ads7950_init_hw (st); -if (ret) { -dev_err (&spi->dev,"Failed to init adc chip\n" ); - goto error_cleanup_ring; - } + error =devm_iio_triggered_buffer_setup (&spi->dev, indio_dev,NULL , + &ti_ads7950_trigger_handler, +NULL ); +if (error) +return dev_err_probe (&spi->dev, error, +"Failed to setup triggered buffer\n" ); - ret =iio_device_register (indio_dev); -if (ret) { -dev_err (&spi->dev,"Failed to register iio device\n" ); - goto error_cleanup_ring; - } + error =ti_ads7950_init_hw (st); +if (error) +return dev_err_probe (&spi->dev, error, +"Failed to init adc chip\n" ); + + error =devm_iio_device_register (&spi->dev, indio_dev); +if (error) +return dev_err_probe (&spi->dev, error, +"Failed to register iio device\n" );/* Add GPIO chip */ st->chip.label =dev_name (&st->spi->dev);
@@ -642,36 +649,12 @@ static int ti_ads7950_probe(struct spi_device *spi) st->chip.get = ti_ads7950_get; st->chip.set = ti_ads7950_set; - ret =gpiochip_add_data (&st->chip, st); -if (ret) { -dev_err (&spi->dev,"Failed to init GPIOs\n" ); - goto error_iio_device; - } + error =devm_gpiochip_add_data (&spi->dev, &st->chip, st); +if (error) +return dev_err_probe (&spi->dev, error, +"Failed to init GPIOs\n" );return 0 ; - -error_iio_device: -iio_device_unregister (indio_dev); -error_cleanup_ring: -iio_triggered_buffer_cleanup (indio_dev); -error_disable_reg: -regulator_disable (st->reg); -error_destroy_mutex: -mutex_destroy (&st->slock); - -return ret; -} - -static void ti_ads7950_remove (struct spi_device *spi) -{ -struct iio_dev *indio_dev =spi_get_drvdata (spi); -struct ti_ads7950_state *st =iio_priv (indio_dev); - -gpiochip_remove (&st->chip); -iio_device_unregister (indio_dev); -iio_triggered_buffer_cleanup (indio_dev); -regulator_disable (st->reg); -mutex_destroy (&st->slock); }static const struct spi_device_id ti_ads7950_id[] = {
@@ -714,7 +697,6 @@ static struct spi_driver ti_ads7950_driver = { .of_match_table = ads7950_of_table, }, .probe = ti_ads7950_probe, - .remove = ti_ads7950_remove, .id_table = ti_ads7950_id, };module_spi_driver (ti_ads7950_driver);
Signature · 1 line
2.53.0.335.g19a08e0c02-goog
Re: [PATCH v2 4/4] iio: adc: ti-ads7950: complete conversion to using managed resources
Quoted 2 lines · Dmitry Torokhov
All resources that the driver needs have managed API now. Switch to using them to make code clearer and drop
ti_ads7950_remove().
...
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_init_hw(struct ti_ads7950_state *st) · line 543
Quoted 7 lines
static int
ti_ads7950_probe(struct spi_device *spi){drivers/iio/adc/ti-ads7950.c +1
-1struct ti_ads7950_state *st;struct iio_dev *indio_dev;const struct ti_ads7950_chip_info *info; -int ret; +int error;
Unrelated change.
...
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_init_hw(struct ti_ads7950_state *st) · line 551
Quoted 10 lines
drivers/iio/adc/ti-ads7950.c +3
-3spi->bits_per_word =16 ; spi->mode |=SPI_CS_WORD ; - ret =spi_setup (spi); -if (ret <0 ) { + error =spi_setup (spi); +if (error) {dev_err (&spi->dev,"Error in spi setup\n" ); -return ret; +return error; }
Ditto.
And since there is already dev_err_probe() in use, I would expect this also be converted.
Would be also nice to use
struct device *dev = &spi->dev;
to make less LoCs and/or make them shorter.
Signature · 2 lines
With Best Regards, Andy Shevchenko
Re: [PATCH v2 4/4] iio: adc: ti-ads7950: complete conversion to using managed resources
Quoted 31 lines · Andy Shevchenko
Quoted 2 lines · Dmitry Torokhov
All resources that the driver needs have managed API now. Switch to using them to make code clearer and drop
ti_ads7950_remove()....
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_init_hw(struct ti_ads7950_state *st) · line 543
Quoted 7 lines
static int
ti_ads7950_probe(struct spi_device *spi){drivers/iio/adc/ti-ads7950.c +1
-1struct ti_ads7950_state *st;struct iio_dev *indio_dev;const struct ti_ads7950_chip_info *info; -int ret; +int error;Unrelated change.
...
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_init_hw(struct ti_ads7950_state *st) · line 551
Quoted 10 lines
drivers/iio/adc/ti-ads7950.c +3
-3spi->bits_per_word =16 ; spi->mode |=SPI_CS_WORD ; - ret =spi_setup (spi); -if (ret <0 ) { + error =spi_setup (spi); +if (error) {dev_err (&spi->dev,"Error in spi setup\n" ); -return ret; +return error; }Ditto.
Quoted 2 lines
And since there is already
dev_err_probe()in use, I would expect this also be converted.
But that would be unrelated change ;)
Quoted 6 lines
Would be also nice to use
struct device *dev = &spi->dev;to make less LoCs and/or make them shorter.
I actually dislike introducing such temporaries in majority of the cases. It makes it hard to follow what device we are dealing with and does not make the code significantly shorter.
Thanks.
Signature · 1 line
Dmitry
Re: [PATCH v2 4/4] iio: adc: ti-ads7950: complete conversion to using managed resources
Quoted 2 lines · Dmitry Torokhov
Quoted 1 line · Andy Shevchenko
On Wed, Feb 18, 2026 at 06:29:28PM -0800, Dmitry Torokhov wrote:
...
Quoted 10 lines
Quoted 8 lines
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_init_hw(struct ti_ads7950_state *st) · line 551
Quoted 5 lines
drivers/iio/adc/ti-ads7950.c +2
-1+if (error) {dev_err (&spi->dev,"Error in spi setup\n" ); -return ret; +return error; }And since there is already
dev_err_probe()in use, I would expect this also be converted.But that would be unrelated change ;)
I never told that this should be done here.
...
Quoted 9 lines
Quoted 5 lines
Would be also nice to use
struct device *dev = &spi->dev;to make less LoCs and/or make them shorter.
I actually dislike introducing such temporaries in majority of the cases. It makes it hard to follow what device we are dealing with and does not make the code significantly shorter.
May be, but my experience is telling me different story. So we have a disagreement here, I leave it to Jonathan to mediate.
Signature · 2 lines
With Best Regards, Andy Shevchenko
Jonathan Cameron
maintainerRe: [PATCH v2 4/4] iio: adc: ti-ads7950: complete conversion to using managed resources
Quoted 27 lines · On Sun, 22 Feb 2026 21:12:17 +0200 Andy Shevchenko…
Quoted 2 lines · Dmitry Torokhov
Quoted 1 line · Andy Shevchenko
On Wed, Feb 18, 2026 at 06:29:28PM -0800, Dmitry Torokhov wrote:
...
Quoted 10 lines
Quoted 8 lines
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_init_hw(struct ti_ads7950_state *st) · line 551
Quoted 5 lines
drivers/iio/adc/ti-ads7950.c +2
-1+if (error) {dev_err (&spi->dev,"Error in spi setup\n" ); -return ret; +return error; }And since there is already
dev_err_probe()in use, I would expect this also be converted.But that would be unrelated change ;)
I never told that this should be done here.
...
Quoted 6 lines
Quoted 5 lines
Would be also nice to use
struct device *dev = &spi->dev;to make less LoCs and/or make them shorter.
Quoted 8 lines
Quoted 3 lines
I actually dislike introducing such temporaries in majority of the cases. It makes it hard to follow what device we are dealing with and does not make the code significantly shorter.
May be, but my experience is telling me different story. So we have a disagreement here, I leave it to Jonathan to mediate.
When there are multiple struct device isntances that we access in the driver I fully agree with Dmitry that it is helpful to fully enumerate them. However, for IIO drivers we keep the one buried in struct iio_dev fairly well hidden so the confusion "opportunity" doesn't tend to occur. As such I tend to come down just on the "have a local variable" side of things.
Jonathan
David Lechner
reviewerRe: [PATCH v2 4/4] iio: adc: ti-ads7950: complete conversion to using managed resources
In reply to Dmitry TorokhovQuoted 13 lines · Dmitry Torokhov
All resources that the driver needs have managed API now. Switch to using them to make code clearer and drop
ti_ads7950_remove().Signed-off-by Dmitry Torokhov <[email protected]>--- drivers/iio/adc/ti-ads7950.c | 98 +++++++++++++++--------------------- 1 file changed, 40 insertions(+), 58 deletions(-)drivers/iio/adc/ti-ads7950.c +0
-0
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_init_hw(struct ti_ads7950_state *st) · line 537
Quoted 10 lines
drivers/iio/adc/ti-ads7950.c +7
-0@@ -528,19 +528,26 @@ static int ti_ads7950_init_hw(struct ti_ads7950_state *st)return 0 ; } +static void ti_ads7950_power_off (void *data) +{ +struct ti_ads7950_state *st = data; + +regulator_disable (st->reg); +} +
For the regulator part, we can simplify this even more.
The part where we call regulator_get_voltage(st->reg); in ti_ads7950_get_range() doesn't actually need to be done there. It was something that I just naively copied from another driver. (This was my first IIO driver after all!)
Instead, we can use devm_regulator_get_enable_read_voltage() in the probe function and just store the voltage in struct ti_ads7950_state.
I would do this in a separate patch first, then the rest of of the devm stuff after that.
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_init_hw(struct ti_ads7950_state *st) · line 543
Quoted 7 lines
drivers/iio/adc/ti-ads7950.c +1
-1static int ti_ads7950_probe (struct spi_device *spi) {struct ti_ads7950_state *st;struct iio_dev *indio_dev;const struct ti_ads7950_chip_info *info; -int ret; +int error;
As in the other patches, please do not rename ret. It is adding noise in the diff. (And I like ret better anyway.)
Jonathan Cameron
maintainerRe: [PATCH v2 4/4] iio: adc: ti-ads7950: complete conversion to using managed resources
In reply to Dmitry TorokhovQuoted 4 lines · On Wed, 18 Feb 2026 18:29:28 -0800 Dmitry Torokhov <[email protected]>…
All resources that the driver needs have managed API now. Switch to using them to make code clearer and drop
ti_ads7950_remove().Signed-off-by Dmitry Torokhov <[email protected]>
Hi Dmitry
One additional comment from me.
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_probe(struct spi_device *spi) · line 608
Quoted 11 lines
static int
ti_ads7950_probe(struct spi_device *spi){drivers/iio/adc/ti-ads7950.c +1
-2indio_dev =devm_iio_device_alloc (&spi->dev,sizeof (*st));@@ -598,36 +605,36 @@ static int ti_ads7950_probe(struct spi_device *spi)mutex_init (&st->slock); st->reg =devm_regulator_get (&spi->dev,"vref" ); -if (IS_ERR (st->reg)) { - ret = dev_err_probe(&spi->dev, PTR_ERR(st->reg), + error =PTR_ERR_OR_ZERO (st->reg);
To me this reads worse than original IS_ERR() / PTR_ERR() pair.
Quoted 15 lines
drivers/iio/adc/ti-ads7950.c +6
-7+if (error) +return dev_err_probe (&spi->dev, error,"Failed to get regulator \"vref\"\n" ); - goto error_destroy_mutex; - } - ret =regulator_enable (st->reg); -if (ret) { -dev_err (&spi->dev,"Failed to enable regulator \"vref\"\n" ); - goto error_destroy_mutex; - } + error =regulator_enable (st->reg); +if (error) +return dev_err_probe (&spi->dev, error, +"Failed to enable regulator \"vref\"\n" );
Re: [PATCH v2 4/4] iio: adc: ti-ads7950: complete conversion to using managed resources
Quoted 19 lines · Jonathan Cameron
Quoted 4 lines · On Wed, 18 Feb 2026 18:29:28 -0800 Dmitry Torokhov <[email protected]>…
All resources that the driver needs have managed API now. Switch to using them to make code clearer and drop
ti_ads7950_remove().Signed-off-by Dmitry Torokhov <[email protected]>Hi Dmitry
One additional comment from me.
Quoted 7 lines
static int
ti_ads7950_probe(struct spi_device *spi){drivers/iio/adc/ti-ads7950.c +0
-0indio_dev =devm_iio_device_alloc (&spi->dev,sizeof (*st));@@ -598,36 +605,36 @@ static int ti_ads7950_probe(struct spi_device *spi)mutex_init (&st->slock);
Quoted 6 lines
drivers/iio/adc/ti-ads7950.c static int ti_ads7950_probe(struct spi_device *spi) · line 608
Quoted 4 lines
drivers/iio/adc/ti-ads7950.c +1
-2st->reg =devm_regulator_get (&spi->dev,"vref" ); -if (IS_ERR (st->reg)) { - ret = dev_err_probe(&spi->dev, PTR_ERR(st->reg), + error =PTR_ERR_OR_ZERO (st->reg);To me this reads worse than original
IS_ERR()/PTR_ERR()pair.
OK, I'll keep that in mind. It is no longer there anyways after converting to devm_regulator_get_enable_read_voltage() that David suggested.
Thanks.
Signature · 1 line
Dmitry