[PATCH dt-schema.git] schemas: add NVMEM cell with #nvmem-cell-cells
zajec5 at gmail.com
Mon Nov 21 03:29:25 PST 2022
On 21.11.2022 12:25, Krzysztof Kozlowski wrote:
> On 21/11/2022 11:58, Rafał Miłecki wrote:
>> From: Rafał Miłecki <rafal at milecki.pl>
>> Some NVMEM cells may provide more than 1 value. An example can be base
>> MAC address that is used for calculating multiple MACs (for few
>> different devices). For specifying value to read phandle needs to be
>> used with an argument.
>> Cc: Srinivas Kandagatla" <srinivas.kandagatla at linaro.org>
>> Cc: Michael Walle <michael at walle.cc>
>> Signed-off-by: Rafał Miłecki <rafal at milecki.pl>
>> This bit is required for moving forward with the
>> [PATCH v2 00/20] nvmem: core: introduce NVMEM layouts
>> As pointed out by Rob #nvmem-cell-cells should be added to the dt-schema
>> and not a Linux binding:
>> Re: [PATCH v2 15/20] dt-bindings: nvmem: add YAML schema for the sl28 vpd layout
>> sl28 is one example that needs #nvmem-cell-cells
>> u-boot,env is another one
>> Please let me know if I got this binding right at all.
>> dtschema/schemas/nvmem/nvmem-cell.yaml | 23 +++++++++++++++++++++++
>> 1 file changed, 23 insertions(+)
>> create mode 100644 dtschema/schemas/nvmem/nvmem-cell.yaml
>> diff --git a/dtschema/schemas/nvmem/nvmem-cell.yaml b/dtschema/schemas/nvmem/nvmem-cell.yaml
>> new file mode 100644
>> index 0000000..dfa99b8
>> --- /dev/null
>> +++ b/dtschema/schemas/nvmem/nvmem-cell.yaml
>> @@ -0,0 +1,23 @@
>> +# SPDX-License-Identifier: BSD-2-Clause
>> +$id: "http://devicetree.org/schemas/nvmem/nvmem-cell.yaml#"
>> +$schema: "http://devicetree.org/meta-schemas/base.yaml#"
> Drop quotes from both.
>> +title: NVMEM Cell Common Properties
>> +description: "Schema for NVMEM cell devicetree bindings"
> Drop quotes.
>> + - Rafał Miłecki <rafal at milecki.pl>
>> + - Srinivas Kandagatla <srinivas.kandagatla at linaro.org>
>> +# always select the core schema
>> +select: true
>> + "#nvmem-cell-cells":
>> + $ref: "/schemas/types.yaml#/definitions/uint32"
> Drop quotes.
> Rest looks fine to me, except a bit funny cell-cell name, but that's
> just life. :)
Bothers me a bit too, I was even thinking about renaming NVMEM cells to
NVMEM entries but I'm not sure if it's not too late for that.
We already have "nvmem-cells" and "nvmem-cell-names" properties so
adding "entries" not would be probably confusing.
>> + description: >
>> + Some NVMEM cells may consist of raw data that requires some extra
>> + processing. They may provide more than 1 value.
>> + Passing extra argument(s) allows selecting data to access.
>> +additionalProperties: true
More information about the linux-mtd