This types of NAND partitions appear in OpenWrt and
U-Boot.
Signed-off-by: Linus Walleij <[email protected]>
---
ChangeLog v1->v2:
- Fix up the binding to be childless
---
.../bindings/mtd/partitions/partitions.yaml | 1 +
.../devicetree/bindings/mtd/partitions/seama.yaml | 50 ++++++++++++++++++++++
2 files changed, 51 insertions(+)
diff --git a/Documentation/devicetree/bindings/mtd/partitions/partitions.yaml b/Documentation/devicetree/bindings/mtd/partitions/partitions.yaml
index 2edc65e0e361..9bd0a8d800da 100644
--- a/Documentation/devicetree/bindings/mtd/partitions/partitions.yaml
+++ b/Documentation/devicetree/bindings/mtd/partitions/partitions.yaml
@@ -21,6 +21,7 @@ oneOf:
- $ref: linksys,ns-partitions.yaml
- $ref: qcom,smem-part.yaml
- $ref: redboot-fis.yaml
+ - $ref: seama.yaml
properties:
compatible: true
diff --git a/Documentation/devicetree/bindings/mtd/partitions/seama.yaml b/Documentation/devicetree/bindings/mtd/partitions/seama.yaml
new file mode 100644
index 000000000000..5ed7764bba69
--- /dev/null
+++ b/Documentation/devicetree/bindings/mtd/partitions/seama.yaml
@@ -0,0 +1,50 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/mtd/partitions/seama.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Seattle Image Partitions
+
+description: The SEAttle iMAge (SEAMA) partition is a type of partition
+ used for NAND flash devices. This type of flash image is found in some
+ D-Link routers such as DIR-645, DIR-842, DIR-859, DIR-860L, DIR-885L,
+ DIR890L and DCH-M225, as well as in WD and NEC routers on the ath79
+ (MIPS), Broadcom BCM53xx, and RAMIPS platforms. This partition type
+ does not have children defined in the device tree, they need to be
+ detected by software.
+
+allOf:
+ - $ref: partition.yaml#
+
+maintainers:
+ - Linus Walleij <[email protected]>
+
+select: false
+
+properties:
+ compatible:
+ const: seama
+
+ '#address-cells': false
+
+ '#size-cells': false
+
+required:
+ - compatible
+
+additionalProperties: false
+
+examples:
+ - |
+ partitions {
+ compatible = "fixed-partitions";
+ #address-cells = <1>;
+ #size-cells = <1>;
+
+ partition@0 {
+ compatible = "seama";
+ reg = <0x0 0x800000>;
+ label = "firmware";
+ };
+ };
--
2.34.1
On Wed, Jul 05, 2023 at 12:17:46PM +0200, Linus Walleij wrote:
> This types of NAND partitions appear in OpenWrt and
> U-Boot.
>
> Signed-off-by: Linus Walleij <[email protected]>
> ---
> ChangeLog v1->v2:
> - Fix up the binding to be childless
> ---
> .../bindings/mtd/partitions/partitions.yaml | 1 +
> .../devicetree/bindings/mtd/partitions/seama.yaml | 50 ++++++++++++++++++++++
> 2 files changed, 51 insertions(+)
>
> diff --git a/Documentation/devicetree/bindings/mtd/partitions/partitions.yaml b/Documentation/devicetree/bindings/mtd/partitions/partitions.yaml
> index 2edc65e0e361..9bd0a8d800da 100644
> --- a/Documentation/devicetree/bindings/mtd/partitions/partitions.yaml
> +++ b/Documentation/devicetree/bindings/mtd/partitions/partitions.yaml
> @@ -21,6 +21,7 @@ oneOf:
> - $ref: linksys,ns-partitions.yaml
> - $ref: qcom,smem-part.yaml
> - $ref: redboot-fis.yaml
> + - $ref: seama.yaml
I think this is in the wrong place. These should be device level
partitioning schemas, not an individual partition type. (Though nesting
is possible)
>
> properties:
> compatible: true
> diff --git a/Documentation/devicetree/bindings/mtd/partitions/seama.yaml b/Documentation/devicetree/bindings/mtd/partitions/seama.yaml
> new file mode 100644
> index 000000000000..5ed7764bba69
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/mtd/partitions/seama.yaml
> @@ -0,0 +1,50 @@
> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/mtd/partitions/seama.yaml#
> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: Seattle Image Partitions
> +
> +description: The SEAttle iMAge (SEAMA) partition is a type of partition
> + used for NAND flash devices. This type of flash image is found in some
> + D-Link routers such as DIR-645, DIR-842, DIR-859, DIR-860L, DIR-885L,
> + DIR890L and DCH-M225, as well as in WD and NEC routers on the ath79
> + (MIPS), Broadcom BCM53xx, and RAMIPS platforms. This partition type
> + does not have children defined in the device tree, they need to be
> + detected by software.
> +
> +allOf:
> + - $ref: partition.yaml#
> +
> +maintainers:
> + - Linus Walleij <[email protected]>
> +
> +select: false
Remove this and your example will fail. You need unevaluatedProperties.
There's a problem in partitions.yaml. It never gets applied, so this
schema never gets applied. The default 'select' is generated based on
$nodename or compatible, but it has neither. This needs some more
thought on how to fix given the variable way partitions can be combined.
Probably at a minimum, all the 'select: false' need to be removed.
> +
> +properties:
> + compatible:
> + const: seama
> +
> + '#address-cells': false
> +
> + '#size-cells': false
> +
> +required:
> + - compatible
> +
> +additionalProperties: false
> +
> +examples:
> + - |
> + partitions {
> + compatible = "fixed-partitions";
> + #address-cells = <1>;
> + #size-cells = <1>;
> +
> + partition@0 {
> + compatible = "seama";
> + reg = <0x0 0x800000>;
> + label = "firmware";
> + };
> + };
>
> --
> 2.34.1
>
On Wed, Jul 5, 2023 at 9:51 PM Rob Herring <[email protected]> wrote:
> On Wed, Jul 05, 2023 at 12:17:46PM +0200, Linus Walleij wrote:
> > - $ref: redboot-fis.yaml
> > + - $ref: seama.yaml
>
> I think this is in the wrong place. These should be device level
> partitioning schemas, not an individual partition type. (Though nesting
> is possible)
OK it wasn't there in v1, then Krzysztof said it should be there
(if I understood correctly) and I will remove it again then.
> > +select: false
>
> Remove this and your example will fail. You need unevaluatedProperties.
OK
> There's a problem in partitions.yaml. It never gets applied, so this
> schema never gets applied. The default 'select' is generated based on
> $nodename or compatible, but it has neither. This needs some more
> thought on how to fix given the variable way partitions can be combined.
> Probably at a minimum, all the 'select: false' need to be removed.
I don't think I can fix this problem, I'm not that good with schema.
I don't even understand the thinking around the existing partitions.yaml.
Yours,
Linus Walleij