On 11/13/25 2:41 PM, Bartosz Golaszewski wrote:

On Thu, Nov 13, 2025 at 2:33 PM Konrad Dybcio
[off-list ref] wrote:

quoted

On 11/13/25 2:18 PM, Bartosz Golaszewski wrote:

quoted

On Fri, Nov 7, 2025 at 11:49 AM Konrad Dybcio
[off-list ref] wrote:

quoted

On 11/7/25 11:29 AM, Bartosz Golaszewski wrote:

quoted

From: Bartosz Golaszewski <redacted>

In order to drop the dependency on CONFIG_OF, convert all device property
getters from OF-specific to generic device properties and stop pulling
in any linux/of.h symbols.

Signed-off-by: Bartosz Golaszewski <redacted>
---

[...]

quoted

-     if (of_property_read_bool(np, "snps,tso"))
+     if (device_property_present(dev, "snps,tso"))

This is a change in behavior - "snps,tso = <0>" would have previously
returned false, it now returns true

This property is a boolean flag, it cannot have a value.

Every DT property may have a value, so this is not as obvious as we'd
like it to be (IIUC - unless that changed recently)

That's new to me. I thought that if a property is a
/schemas/types.yaml#/definitions/flag then only its boolean form is
allowed. The fact that the dtc can compile it with a value doesn't
matter as the bindings are the higher authority for DT sources?

You're right, I misrepresented the issue. read_bool() was historically
used for making sure a property is present and this was a problem
when the type was != flag.

Please discard my comment

Konrad