在 2023/6/1 上午4:06, Krzysztof Kozlowski 写道:
On 25/05/2023 04:22, zhuyinbo wrote:
在 2023/5/24 下午6:29, Conor Dooley 写道:
On Wed, May 24, 2023 at 05:44:38PM +0800, zhuyinbo wrote:
在 2023/5/24 下午4:56, Conor Dooley 写道:
On Mon, May 22, 2023 at 03:10:29PM +0800, Yinbo Zhu wrote:
Add the Loongson platform spi binding with DT schema format using
json-schema.
Signed-off-by: Yinbo Zhu <zhuyinbo@xxxxxxxxxxx>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx>
---
.../bindings/spi/loongson,ls2k-spi.yaml | 41 +++++++++++++++++++
MAINTAINERS | 6 +++
2 files changed, 47 insertions(+)
create mode 100644 Documentation/devicetree/bindings/spi/loongson,ls2k-spi.yaml
diff --git a/Documentation/devicetree/bindings/spi/loongson,ls2k-spi.yaml b/Documentation/devicetree/bindings/spi/loongson,ls2k-spi.yaml
new file mode 100644
index 000000000000..d0be6e5378d7
--- /dev/null
+++ b/Documentation/devicetree/bindings/spi/loongson,ls2k-spi.yaml
@@ -0,0 +1,41 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/spi/loongson,ls2k-spi.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Loongson SPI controller
+
+maintainers:
+ - Yinbo Zhu <zhuyinbo@xxxxxxxxxxx>
+
+allOf:
+ - $ref: /schemas/spi/spi-controller.yaml#
+
+properties:
+ compatible:
+ enum:
+ - loongson,ls2k-spi
I am sorry to jump in here at such a late stage with a (potentially)
trivial question. "ls2k" is the SoC family rather than a specific model
as far as I understand.
The answer is probably yes, but do all SoCs in the family have an
identical version of the IP?
No, but the spi supported by this loongson spi driver are all the same
identical version, and other type or verion spi will be supported as
needed in the future.
Does having a catch-all compatible make sense then when not all SoCs in
the ls2k family will actually be able to use this driver?
Yes, it is make sense as it can reduce the workload of the community.
I missed it - that's a great comment. Hm, I don't know... Reviewing
Loongson patches is quite a work, so I don't see here reduced workload.
If we do not consider the differences in SPI hardware and consider the
differences in SoC, it will result for each new a SoC adaptation, a new
compatible patch needs to be submitted to the community but spi hardware
was same and that will increase the workload of the community, It seems
to be more appropriate that use same compatible when spi hardware was
same and use different compatible when spi hardware was different.
Please read existing guidelines:
https://elixir.bootlin.com/linux/v6.1-rc1/source/Documentation/devicetree/bindings/writing-bindings.rst >
All of them.
okay, I got it.
Thanks.
