[PATCH] dt-bindings: mfd: ti,bq25703a: Expand to include BQ25798

[PATCH] dt-bindings: mfd: ti,bq25703a: Expand to include BQ25798

[Alexey Charkov]

TI has unpublished the BQ25792 charger variant and pulled its docs from
their website.

BQ25798 is a pin and register compatible replacement, which also adds MPPT
tracking support (reg 0x15, added in BQ25798 and reserved in BQ25792) and
backup mode (reg 0x0f[1], reg 0x10[7:6], reg 0x16[0] and reg 0x1c[7:5] on
BQ25798 - all of them reserved on BQ25792).

Expand the existing binding to include BQ25798 with a fallback compatible
to BQ25792.

Link: https://www.ti.com/product/BQ25798
Signed-off-by: Alexey Charkov <alchark@flipper.net>
---
 Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml b/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml
index cdce83f05804..43afd232a614 100644
--- a/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml
+++ b/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml
@@ -11,9 +11,14 @@ maintainers:
 
 properties:
   compatible:
-    enum:
-      - ti,bq25703a
-      - ti,bq25792
+    oneOf:
+      - enum:
+          - ti,bq25703a
+          - ti,bq25792
+      - items:
+          - enum:
+              - ti,bq25798
+          - const: ti,bq25792
 
   reg:
     const: 0x6b

---
base-commit: 68142f986ff04b2b70b31db00f719bf690f64a9a
change-id: 20260914-bq25798-67768a40517e

Best regards,
--  
Alexey Charkov <alchark@flipper.net>

Re: [PATCH] dt-bindings: mfd: ti,bq25703a: Expand to include BQ25798

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Adding an array-based fallback compatible string without updating the `if: properties: compatible: const:` conditionals causes the new compatible to bypass device-specific schema constraints.
--

commit 37d04d6cb52639a3cc4ba9ade1878fa540a5cb3e
Author: Alexey Charkov <alchark@flipper.net>

dt-bindings: mfd: ti,bq25703a: Expand to include BQ25798

TI has unpublished the BQ25792 charger variant and pulled its docs from
their website. BQ25798 is a pin and register compatible replacement. Expand
the existing binding to include BQ25798 with a fallback compatible to
BQ25792.

Link: https://www.ti.com/product/BQ25798

> diff --git a/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml b/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml
> index cdce83f058041..43afd232a6142 100644
> --- a/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml
> +++ b/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml
> @@ -11,9 +11,14 @@ maintainers:
>  
>  properties:
>    compatible:
> -    enum:
> -      - ti,bq25703a
> -      - ti,bq25792
> +    oneOf:
> +      - enum:
> +          - ti,bq25703a
> +          - ti,bq25792
> +      - items:
> +          - enum:
> +              - ti,bq25798
> +          - const: ti,bq25792

[Severity: Medium]
Is this a regression that causes the new compatible string to bypass
device-specific schema constraints?

Later in the file's allOf/if section, there is an exact match condition
for the ti,bq25792 compatible string:

  - if:
      properties:
        compatible:
          const: ti,bq25792

Since the schema const keyword mandates strict equality, it will fail to
match the fallback array ["ti,bq25798", "ti,bq25792"]. As a result, if a
devicetree author specifies compatible = "ti,bq25798", "ti,bq25792" and
configures out-of-bounds regulator or current values, the device-specific
hardware constraints for ti,bq25792 will not be applied during validation.

Could this if condition be updated to use contains instead of const, or
otherwise include the new fallback array?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260914-bq25798-v1-1-dc5e2ec343c4@flipper.net?part=1

Re: [PATCH] dt-bindings: mfd: ti,bq25703a: Expand to include BQ25798

[Alexey Charkov]

On Mon, Sep 14, 2026 at 6:08 PM <sashiko-bot@kernel.org> wrote:
>
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [Medium] Adding an array-based fallback compatible string without updating the `if: properties: compatible: const:` conditionals causes the new compatible to bypass device-specific schema constraints.
> --
>
> commit 37d04d6cb52639a3cc4ba9ade1878fa540a5cb3e
> Author: Alexey Charkov <alchark@flipper.net>
>
> dt-bindings: mfd: ti,bq25703a: Expand to include BQ25798
>
> TI has unpublished the BQ25792 charger variant and pulled its docs from
> their website. BQ25798 is a pin and register compatible replacement. Expand
> the existing binding to include BQ25798 with a fallback compatible to
> BQ25792.
>
> Link: https://www.ti.com/product/BQ25798
>
> > diff --git a/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml b/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml
> > index cdce83f058041..43afd232a6142 100644
> > --- a/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml
> > +++ b/Documentation/devicetree/bindings/mfd/ti,bq25703a.yaml
> > @@ -11,9 +11,14 @@ maintainers:
> >
> >  properties:
> >    compatible:
> > -    enum:
> > -      - ti,bq25703a
> > -      - ti,bq25792
> > +    oneOf:
> > +      - enum:
> > +          - ti,bq25703a
> > +          - ti,bq25792
> > +      - items:
> > +          - enum:
> > +              - ti,bq25798
> > +          - const: ti,bq25792
>
> [Severity: Medium]
> Is this a regression that causes the new compatible string to bypass
> device-specific schema constraints?
>
> Later in the file's allOf/if section, there is an exact match condition
> for the ti,bq25792 compatible string:
>
>   - if:
>       properties:
>         compatible:
>           const: ti,bq25792
>
> Since the schema const keyword mandates strict equality, it will fail to
> match the fallback array ["ti,bq25798", "ti,bq25792"]. As a result, if a
> devicetree author specifies compatible = "ti,bq25798", "ti,bq25792" and
> configures out-of-bounds regulator or current values, the device-specific
> hardware constraints for ti,bq25792 will not be applied during validation.
>
> Could this if condition be updated to use contains instead of const, or
> otherwise include the new fallback array?

Fair point, will address in v2.

Best regards,
Alexey