The device powers up in I2C mode. Switching to SPI mode
requires sending a sequence of SPI writes as described in
the datasheet.

During this sequence, the device may still be in I2C mode,
so SPI transactions may not be recognized and can fail.
Such errors are expected and are ignored.

Add a comment to clarify this behavior.

Signed-off-by: Hungyu Lin <dennylin0707@gmail.com>
---
Changes in v2:
- Add datasheet reference
- Add TL;DR note

 drivers/staging/iio/addac/adt7316-spi.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/staging/iio/addac/adt7316-spi.c b/drivers/staging/iio/addac/adt7316-spi.c
index f91325d11394..980532b7fa1e 100644
--- a/drivers/staging/iio/addac/adt7316-spi.c
+++ b/drivers/staging/iio/addac/adt7316-spi.c
@@ -106,7 +106,17 @@ static int adt7316_spi_probe(struct spi_device *spi_dev)
 		return -EINVAL;
 	}
 
-	/* switch from default I2C protocol to SPI protocol */
+	/*
+	 * The device powers up in I2C mode. Switching to SPI mode
+	 * requires sending a sequence of SPI writes as described in
+	 * the datasheet ("Serial Interface Selection" section).
+	 *
+	 * During this sequence, the device may still be in I2C mode,
+	 * so SPI transactions may not be recognized and can fail.
+	 * Such errors are expected and are ignored.
+	 *
+	 * TL;DR: Do not change this!
+	 */
 	adt7316_spi_write(spi_dev, 0, 0);
 	adt7316_spi_write(spi_dev, 0, 0);
 	adt7316_spi_write(spi_dev, 0, 0);
-- 
2.34.1

On Sun, May 10, 2026 at 08:10:41AM +0000, Hungyu Lin wrote:
> The device powers up in I2C mode. Switching to SPI mode
> requires sending a sequence of SPI writes as described in
> the datasheet.
> 
> During this sequence, the device may still be in I2C mode,
> so SPI transactions may not be recognized and can fail.
> Such errors are expected and are ignored.
> 
> Add a comment to clarify this behavior.

...

> Changes in v2:
> - Add datasheet reference

This is still odd. Please, put in a format

	$SNUM "$STITLE"

where $SNUM is the section number (like 2.1.4) and "$STITLE" as currently done
in this version, Ideally the datasheet also needs a versioning.

...

With that being addressed,
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>

-- 
With Best Regards,
Andy Shevchenko

> This is still odd. Please, put in a format
>
> $SNUM "$STITLE"
>
> where $SNUM is the section number (like 2.1.4) and "$STITLE" as currently done
> in this version, Ideally the datasheet also needs a versioning.

The ADT7316 datasheet does not seem to provide explicit section
numbering for the "Serial Interface Selection" section, only the
section title.

Would it be acceptable to reference it using just the section
title, or would you prefer using the page number instead?

Thanks,
Hungyu

On Sun, May 10, 2026 at 02:22:02AM -0700, Denny Lin wrote:
> > This is still odd. Please, put in a format
> >
> > $SNUM "$STITLE"
> >
> > where $SNUM is the section number (like 2.1.4) and "$STITLE" as currently done
> > in this version, Ideally the datasheet also needs a versioning.
> 
> The ADT7316 datasheet does not seem to provide explicit section
> numbering for the "Serial Interface Selection" section, only the
> section title.
> 
> Would it be acceptable to reference it using just the section
> title, or would you prefer using the page number instead?

Do you have the Datasheet version? Maybe URL?

(Yeah, that's unfortunate they do not provide numbering, in such a case
 it's acceptable to have reduced reference.)

-- 
With Best Regards,
Andy Shevchenko

> Do you have the Datasheet version? Maybe URL?

The datasheet I am referring to is:
https://www.analog.com/en/products/adt7316.html

(ADT7316/ADT7317/ADT7318, Rev. B)

Thanks,
Hungyu

On Sun, May 10, 2026 at 02:31:31AM -0700, Denny Lin wrote:
> > Do you have the Datasheet version? Maybe URL?
> 
> The datasheet I am referring to is:
> https://www.analog.com/en/products/adt7316.html
> 
> (ADT7316/ADT7317/ADT7318, Rev. B)

OK! Put the URL in the commit message as Datasheet tag and use the name in the
comment in the code

Datasheet: $URL
Signed-off-by: ...

...

 /*
  *...datasheet "ADT7316/ADT7317/ADT7318", Rev. B...
  */

-- 
With Best Regards,
Andy Shevchenko

On Sun, May 10, 2026 at 3:11 AM Hungyu Lin <dennylin0707@gmail.com> wrote:
>
> The device powers up in I2C mode. Switching to SPI mode
> requires sending a sequence of SPI writes as described in
> the datasheet.
>
> During this sequence, the device may still be in I2C mode,
> so SPI transactions may not be recognized and can fail.
> Such errors are expected and are ignored.
>
> Add a comment to clarify this behavior.
>
> Signed-off-by: Hungyu Lin <dennylin0707@gmail.com>
> ---
> Changes in v2:
> - Add datasheet reference
> - Add TL;DR note
>
>  drivers/staging/iio/addac/adt7316-spi.c | 12 +++++++++++-
>  1 file changed, 11 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/staging/iio/addac/adt7316-spi.c b/drivers/staging/iio/addac/adt7316-spi.c
> index f91325d11394..980532b7fa1e 100644
> --- a/drivers/staging/iio/addac/adt7316-spi.c
> +++ b/drivers/staging/iio/addac/adt7316-spi.c
> @@ -106,7 +106,17 @@ static int adt7316_spi_probe(struct spi_device *spi_dev)
>                 return -EINVAL;
>         }
>
> -       /* switch from default I2C protocol to SPI protocol */
> +       /*
> +        * The device powers up in I2C mode. Switching to SPI mode
> +        * requires sending a sequence of SPI writes as described in
> +        * the datasheet ("Serial Interface Selection" section).
> +        *
> +        * During this sequence, the device may still be in I2C mode,
> +        * so SPI transactions may not be recognized and can fail.
> +        * Such errors are expected and are ignored.
> +        *
> +        * TL;DR: Do not change this!
> +        */
>         adt7316_spi_write(spi_dev, 0, 0);
>         adt7316_spi_write(spi_dev, 0, 0);
>         adt7316_spi_write(spi_dev, 0, 0);
> --
> 2.34.1
>
>

Reviewed-by: Maxwell Doose <m32285159@gmail.com>

but please, wait at least 24hrs before submitting a new version...

best regards,
max