Subject: Re: [PATCH v2 1/2] mmc: dt-bindings: sdhci-of-aspeed: Add AST2700
fallback compatibles

On Tue, Mar 17, 2026 at 01:56:42PM +0800, Ryan Chen wrote:

quoted

Describe AST2700 as compatible with the existing AST2600 SD controller
and SDHCI bindings by requiring fallback compatibles in the device tree.

Also require `resets` on the AST2700 SD controller node.

Signed-off-by: Ryan Chen <ryan_chen@aspeedtech.com>
---
Changes in v2:
- add missing blank line
- modify ast2700 compatible items const

Why?

quoted

---
 .../devicetree/bindings/mmc/aspeed,sdhci.yaml      | 39

+++++++++++++++++-----

quoted

 1 file changed, 31 insertions(+), 8 deletions(-)

diff --git a/Documentation/devicetree/bindings/mmc/aspeed,sdhci.yaml

b/Documentation/devicetree/bindings/mmc/aspeed,sdhci.yaml
index d24950ccea95..9c8e068964a1 100644

--- a/Documentation/devicetree/bindings/mmc/aspeed,sdhci.yaml
+++ b/Documentation/devicetree/bindings/mmc/aspeed,sdhci.yaml

@@ -22,10 +22,14 @@ description: |+

 properties:
   compatible:
-    enum:
-      - aspeed,ast2400-sd-controller
-      - aspeed,ast2500-sd-controller
-      - aspeed,ast2600-sd-controller
+    oneOf:
+      - const: aspeed,ast2400-sd-controller
+      - const: aspeed,ast2500-sd-controller
+      - const: aspeed,ast2600-sd-controller

No, previous code was correct.

Is this some microslop LLM product? I questioned style last time and now we
got random changes without explanation.

Please confirm - did you use any LLM microslop tools to create this patch?

No I create it with B4. And commit each by each, I don't use LLM to porting this.

Please also confirm - who internally reviewed this patch before posting?

I port this and do test in my side and confirm it.


In https://lore.kernel.org/all/20260314-flat-topaz-peacock-440a9c@quoll/#t (local) 
You request drop driver compatible; 
Base on your instruction, So I do my study in Linux.
Use following to do compatible. Not add driver compatible string.
https://lore.kernel.org/all/TY2PPF5CB9A1BE642DFDE4862B40BBAA1DBF240A@TY2PPF5CB9A1BE6.apcprd06.prod.outlook.com/ (local)
That is my follow up porting, not use LLM.