On 08/03/2017 04:16 AM, Guochun Mao wrote:

Hi Cyrille,

On Wed, 2017-08-02 at 17:31 +0200, Cyrille Pitchen wrote:

quoted

Hi Guochun,

Le 02/08/2017 ? 10:36, Boris Brezillon a ?crit :

quoted

Hi Matthias,

On Tue, 1 Aug 2017 14:33:22 +0200
Matthias Brugger [off-list ref] wrote:

quoted

On 07/25/2017 04:10 AM, Guochun Mao wrote:

quoted

Add "mediatek,mt2712-nor" and "mediatek,mt7622-nor"
for nor flash node's compatible strings.

Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Signed-off-by: Guochun Mao <redacted>
---
   .../devicetree/bindings/mtd/mtk-quadspi.txt        |    2 ++
   1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/mtd/mtk-quadspi.txt b/Documentation/devicetree/bindings/mtd/mtk-quadspi.txt
index 5ded66a..9845ff4 100644
--- a/Documentation/devicetree/bindings/mtd/mtk-quadspi.txt
+++ b/Documentation/devicetree/bindings/mtd/mtk-quadspi.txt

@@ -3,6 +3,8 @@
   Required properties:
   - compatible: 	  The possible values are:
   		  "mediatek,mt2701-nor"
+		  "mediatek,mt2712-nor"
+		  "mediatek,mt7622-nor"
   		  "mediatek,mt7623-nor"
   		  "mediatek,mt8173-nor"
   		  For mt8173, compatible should be "mediatek,mt8173-nor".

There should be another patch to update the mtk_nor_of_ids[] array in
the mtk-quadspi.c driver, shouldn't it?

Currently, I only see the "mediatek,mt8173-nor" compatible string at the
driver side. The driver and its documentation should be aligned and both
patches should be applied at the same time.

quoted

Best regards,

Cyrille

 From my designer's info, these IPs' nor controller could use same
driver, and I had verified it. I think there's no need to update
mtk_nor_of_ids[], unless we find difference between these IPs(exactly
compared with mt8173).
How do you think?

Right, mt8173-nor is the fallback binding which will be used by the driver. As 
the documentation states, the compatible string should be something like:
"mediatek,mt7622-nor", "mediatek,mt8173-nor"

So the mt7622-nor binding is not matched in the driver, as the driver is the 
same on all SoCs. Nonetheless if we realize in the future that mt7622 IP has 
some special feature/bug, we can use the binding to implement that in the driver.

AFAIK this is the standard way for bindings. Cyrille, is this ok for you, or do 
want all bindings to be reflected in the driver even though they don't change 
any functionality?

Regards,
Matthias

And, we will add dts patches for new SOCs, are these patches should
applied with this documentation?

Best regards,
Guochun