Re: [PATCH v3 0/5] iio: dac: ad5504: bindings, cleanups, and scale fixes

[David Lechner <dlechner@baylibre.com>]

On 5/9/26 9:20 AM, Taha Ed-Dafili wrote:
> Hi everyone,
> 
> First, I want to sincerely apologize for the long delay between v2 and v3.
> I had to step away to focus on finishing my last year studies, but I am
> back now to get this series over the finish line!
> 
> This v3 series addresses the feedbacks from the v2 review,
> specifically focusing on strict DT validation, dropping unnecessary
> driver code, and making sure the driver works perfectly at every
> individual commit.
> 
> Testing & Hardware Proof:
> The series has been physically verified on an ARM64 Raspberry Pi 5 using
> a custom Device Tree overlay.
> 
> The AD5504 is a 12-bit DAC. The hardware scale outputs
> now perfectly match the DT inputs:
> 
> Testing the 60V path (output-range-microvolt = <0 60000000>):
> $ cat /sys/bus/iio/devices/iio:device0/out_voltage_scale
> 14.648437500  (60000mV / 4096)
> 
> Testing the 30V path (output-range-microvolt = <0 30000000>):
> $ cat /sys/bus/iio/devices/iio:device0/out_voltage_scale
> 7.324218750   (30000mV / 4096)
> 
> Additionally, providing invalid DT configurations (e.g., 70V) successfully
> triggers the strict bounds checking and cleanly aborts the probe with
> -EINVAL.
> 
> Changes in v3:
> 
> * Dropped the patch adding GPIO control for /CLR and /LDAC. As pointed out
>   by Nuno Sá, since the driver does not actively handle or toggle these
>   pins, requesting them in the driver is dead code. They remain documented
>   in the bindings.
> * Combined the pdata removal, regulator _enable swap, and DT parsing into
>   a single atomic commit (Patch 5) so the driver compiles and runs cleanly
>   at every step of the git history.

It should be possible to do this while still having separate commits.

> * Restored the -ENODEV check for the regulator to maintain ACPI
>   compatibility.

While you were studying we discussed this some more in similar cases
and decided that we would like to handle this differently going forward.
For example, see recent changes in adc/ti-ads7950.c which also has
an ACPI case.

> * Implemented strict min/max array bounds validation for the
>   'output-range-microvolt' property.
> * Included missing <linux/array_size.h>, <asm/byteorder.h>, and
>   <linux/kstrtox.h> headers for full IWYU compliance.
> * Reordered commits to group cleanups before functional changes.
> * Dropped Suggested-by tags for standard review feedback.
>