linux-kernel - Re: [RFC PATCH v6 1/2] dt-bindings: PCI: dw: rockchip: Add rk3576 support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20250221180930.GA3486559-robh@kernel.org>
Date: Fri, 21 Feb 2025 12:09:30 -0600
From: Rob Herring <robh@...nel.org>
To: Kever Yang <kever.yang@...k-chips.com>
Cc: heiko@...ech.de, linux-rockchip@...ts.infradead.org,
	Simon Xue <xxm@...k-chips.com>, Conor Dooley <conor+dt@...nel.org>,
	Bjorn Helgaas <bhelgaas@...gle.com>, linux-pci@...r.kernel.org,
	Krzysztof Wilczyński <kw@...ux.com>,
	linux-kernel@...r.kernel.org,
	Krzysztof Kozlowski <krzk+dt@...nel.org>,
	devicetree@...r.kernel.org,
	Lorenzo Pieralisi <lpieralisi@...nel.org>,
	Shawn Lin <shawn.lin@...k-chips.com>,
	Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>,
	linux-arm-kernel@...ts.infradead.org
Subject: Re: [RFC PATCH v6 1/2] dt-bindings: PCI: dw: rockchip: Add rk3576
 support

On Fri, Feb 21, 2025 at 06:43:56PM +0800, Kever Yang wrote:
> rk3576 is using dwc controller, with msi interrupt directly to gic instead
> of to gic its, so
> - no its support is required and the 'msi-map' is not need anymore,
> - a new 'msi' interrupt is needed.
> 
> Signed-off-by: Kever Yang <kever.yang@...k-chips.com>
> ---
> 
> Changes in v6:
> - Fix make dt_binding_check and make CHECK_DTBS=y
> 
> Changes in v5:
> - Add constraints per device for interrupt-names due to the interrupt is
> different from rk3588.
> 
> Changes in v4:
> - Fix wrong indentation in dt_binding_check report by Rob
> 
> Changes in v3:
> - Fix dtb check broken on rk3588
> - Update commit message
> 
> Changes in v2:
> - remove required 'msi-map'
> - add interrupt name 'msi'
> 
>  .../bindings/pci/rockchip-dw-pcie-common.yaml | 59 +++++++++++++++----
>  .../bindings/pci/rockchip-dw-pcie.yaml        |  4 +-
>  2 files changed, 50 insertions(+), 13 deletions(-)
> 
> diff --git a/Documentation/devicetree/bindings/pci/rockchip-dw-pcie-common.yaml b/Documentation/devicetree/bindings/pci/rockchip-dw-pcie-common.yaml
> index cc9adfc7611c..069eb267c0bb 100644
> --- a/Documentation/devicetree/bindings/pci/rockchip-dw-pcie-common.yaml
> +++ b/Documentation/devicetree/bindings/pci/rockchip-dw-pcie-common.yaml
> @@ -64,6 +64,10 @@ properties:
>            interrupts - aer_rc_err, aer_rc_err_msi, rx_cpl_timeout,
>            tx_cpl_timeout, cor_err_sent, nf_err_sent, f_err_sent, cor_err_rx,
>            nf_err_rx, f_err_rx, radm_qoverflow
> +      - description:
> +          Combinded MSI line interrupt, which is to support MSI interrupts
> +          output to GIC controller via GIC SPI interrupt instead of GIC its
> +          interrupt.
>        - description:
>            eDMA write channel 0 interrupt
>        - description:
> @@ -75,16 +79,6 @@ properties:
>  
>    interrupt-names:
>      minItems: 5
> -    items:

You just made the max 5 by dropping this.

> -      - const: sys
> -      - const: pmc
> -      - const: msg
> -      - const: legacy
> -      - const: err
> -      - const: dma0

Make this 'enum: [ dma0, msi ]'

Yeah, that would allow the wrong one to be used, but I prefer that over 
duplicating the names.

> -      - const: dma1
> -      - const: dma2
> -      - const: dma3
>  
>    num-lanes: true
>  
> @@ -123,4 +117,49 @@ required:
>  
>  additionalProperties: true
>  
> +anyOf:

allOf

> +  - if:
> +      properties:
> +        compatible:
> +          contains:
> +            enum:
> +              - rockchip,rk3568-pcie
> +              - rockchip,rk3588-pcie
> +              - rockchip,rk3588-pcie-ep
> +    then:
> +      properties:
> +        interrupt-names:
> +          minItems: 5

That's already the min.

> +          type: array

Don't need type. Do you see any schema use 'type: array' outside of the 
core? 

> +          items:
> +            - const: sys
> +            - const: pmc
> +            - const: msg
> +            - const: legacy
> +            - const: err
> +            - const: dma0
> +            - const: dma1
> +            - const: dma2
> +            - const: dma3

The whole if/then can be dropped.

> +
> +  - if:
> +      properties:
> +        compatible:
> +          contains:
> +            enum:
> +              - rockchip,rk3576-pcie
> +    then:
> +      properties:
> +        interrupt-names:
> +          type: array
> +          items:
> +            - const: sys
> +            - const: pmc
> +            - const: msg
> +            - const: legacy
> +            - const: err
> +            - const: msi

Don't need type or items. Just this is enough to ensure 'msi' is used:

contains:
  const: msi

> +          minItems: 6
> +          maxItems: 6

This part is correct.

> +
>  ...
> diff --git a/Documentation/devicetree/bindings/pci/rockchip-dw-pcie.yaml b/Documentation/devicetree/bindings/pci/rockchip-dw-pcie.yaml
> index 550d8a684af3..9a464731fa4a 100644
> --- a/Documentation/devicetree/bindings/pci/rockchip-dw-pcie.yaml
> +++ b/Documentation/devicetree/bindings/pci/rockchip-dw-pcie.yaml
> @@ -26,6 +26,7 @@ properties:
>        - const: rockchip,rk3568-pcie
>        - items:
>            - enum:
> +              - rockchip,rk3576-pcie
>                - rockchip,rk3588-pcie
>            - const: rockchip,rk3568-pcie
>  
> @@ -71,9 +72,6 @@ properties:
>  
>    vpcie3v3-supply: true
>  
> -required:
> -  - msi-map

This should be moved to an if/then schema, not just dropped. Unless it 
was wrong for the existing users. If so, then that's a separate patch.

> -
>  unevaluatedProperties: false
>  
>  examples:
> -- 
> 2.25.1
>