On Fri, 15 Jul 2022 14:28:53 +0200
Nuno Sá <nuno.sa@xxxxxxxxxx> wrote:
> This moves the IIO in kernel interface to use fwnode properties and thus
> be firmware agnostic.
>
> Note that the interface is still not firmware agnostic. At this point we
> have both OF and fwnode interfaces so that we don't break any user. On
> top of this we also want to have a per driver conversion and that is the
> main reason we have both of_xlate() and fwnode_xlate() support.
>
> Signed-off-by: Nuno Sá <nuno.sa@xxxxxxxxxx>
> Reviewed-by: Andy Shevchenko <andy.shevchenko@xxxxxxxxx>
Nice. One trivial follow through of wrong comment syntax. I'll fix up
if I pick the series up before you do a respin for other reasons.
> ---
> drivers/iio/inkern.c | 159 ++++++++++++++++++-----------------
> include/linux/iio/consumer.h | 36 ++++----
> include/linux/iio/iio.h | 5 ++
> 3 files changed, 108 insertions(+), 92 deletions(-)
>
> -struct iio_channel *__of_iio_channel_get_by_name(struct device_node *np,
> - const char *name)
> +struct iio_channel *
> +__fwnode_iio_channel_get_by_name(struct fwnode_handle *fwnode, const char *name)
> {
> struct iio_channel *chan;
> int index = 0;
> @@ -220,19 +236,20 @@ struct iio_channel *__of_iio_channel_get_by_name(struct device_node *np,
> /*
> * For named iio channels, first look up the name in the
> * "io-channel-names" property. If it cannot be found, the
> - * index will be an error code, and of_iio_channel_get()
> + * index will be an error code, and fwnode_iio_channel_get()
> * will fail.
> */
> if (name)
> - index = of_property_match_string(np, "io-channel-names", name);
> + index = fwnode_property_match_string(fwnode, "io-channel-names",
> + name);
>
> - chan = of_iio_channel_get(np, index);
> + chan = fwnode_iio_channel_get(fwnode, index);
> if (!IS_ERR(chan) || PTR_ERR(chan) == -EPROBE_DEFER)
> return chan;
> if (name) {
> if (index >= 0) {
> - pr_err("ERROR: could not get IIO channel %pOF:%s(%i)\n",
> - np, name, index);
> + pr_err("ERROR: could not get IIO channel %pfw:%s(%i)\n",
> + fwnode, name, index);
> /*
> * In this case, we found 'name' in 'io-channel-names'
> * but somehow we still fail so that we should not proceed
> @@ -242,16 +259,16 @@ struct iio_channel *__of_iio_channel_get_by_name(struct device_node *np,
> */
> return ERR_PTR(-EINVAL);
> }
> - /* If index < 0, then of_parse_phandle_with_args() fails
> - * with -EINVAL which is expected. We should not proceed
> - * if we get any other error.
Wrong comment syntax. I guess I can fix this one as well if nothing else comes up.
Or I might be lazy and only fix this one as it replaces the previous wrong one.
> + /* If index < 0, then fwnode_property_get_reference_args() fails
> + * with -EINVAL or -ENOENT (ACPI case) which is expected. We
> + * should not proceed if we get any other error.
> */
> - if (PTR_ERR(chan) != -EINVAL)
> + if (PTR_ERR(chan) != -EINVAL && PTR_ERR(chan) != -ENOENT)
> return chan;
> } else if (PTR_ERR(chan) != -ENOENT) {
> /*
> * if !name, then we should only proceed the lookup if
> - * of_parse_phandle_with_args() returns -ENOENT.
> + * fwnode_property_get_reference_args() returns -ENOENT.
> */
> return chan;
> }
> @@ -260,13 +277,14 @@ struct iio_channel *__of_iio_channel_get_by_name(struct device_node *np,
> return ERR_PTR(-ENODEV);
> }
