mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Fenglin Wu <quic_fenglinw@quicinc.com>
To: Rob Herring <robh@kernel.org>
Cc: <linux-arm-msm@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
	<sboyd@kernel.org>, Andy Gross <agross@kernel.org>,
	Bjorn Andersson <bjorn.andersson@linaro.org>,
	Subbaraman Narayanamurthy <quic_subbaram@quicinc.com>,
	<devicetree@vger.kernel.org>, <collinsd@codeaurora.org>,
	<subbaram@codeaurora.org>, <tglx@linutronix.de>, <maz@kernel.org>
Subject: Re: [PATCH v4 11/11] dt-bindings: convert qcom,spmi-pmic-arb binding to YAML format
Date: Fri, 7 Jan 2022 15:03:14 +0800	[thread overview]
Message-ID: <5ef25022-b53b-4002-7de8-e31bfb236009@quicinc.com> (raw)
In-Reply-To: <b35d0f12-1de2-9e15-2d87-5049614eeff1@quicinc.com>



On 2021/12/22 8:45, Fenglin Wu wrote:
> resend with plain text
> 
> 
> On 2021/12/21 22:42, Rob Herring wrote:
>> On Tue, Dec 21, 2021 at 03:20:09PM +0800, Fenglin Wu wrote:
>>> Convert the SPMI PMIC arbiter documentation to JSON/yaml.
>>>
>>> Signed-off-by: Fenglin Wu <quic_fenglinw@quicinc.com>
>>> ---
>>>   .../bindings/spmi/qcom,spmi-pmic-arb.txt           |  67 ----------
>>>   .../bindings/spmi/qcom,spmi-pmic-arb.yaml          | 146 
>>> +++++++++++++++++++++
>>>   2 files changed, 146 insertions(+), 67 deletions(-)
>>>   delete mode 100644 
>>> Documentation/devicetree/bindings/spmi/qcom,spmi-pmic-arb.txt
>>>   create mode 100644 
>>> Documentation/devicetree/bindings/spmi/qcom,spmi-pmic-arb.yaml
>>>
>>
>>> diff --git 
>>> a/Documentation/devicetree/bindings/spmi/qcom,spmi-pmic-arb.yaml 
>>> b/Documentation/devicetree/bindings/spmi/qcom,spmi-pmic-arb.yaml
>>> new file mode 100644
>>> index 0000000..df8cfb7
>>> --- /dev/null
>>> +++ b/Documentation/devicetree/bindings/spmi/qcom,spmi-pmic-arb.yaml
>>> @@ -0,0 +1,146 @@
>>> +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
>>> +%YAML 1.2
>>> +---
>>> +$id: http://devicetree.org/schemas/spmi/qcom,spmi-pmic-arb.yaml#
>>> +$schema: http://devicetree.org/meta-schemas/core.yaml#
>>> +
>>> +title: Qualcomm SPMI PMIC Arbiter
>>> +
>>> +maintainers:
>>> +  - Fenglin Wu <quic_fenglinw@quicinc.com>
>>> +  - Subbaraman Narayanamurthy <quic_subbaram@quicinc.com>
>>> +
>>> +description: |
>>> +  The SPMI PMIC Arbiter is found on Snapdragon chipsets. It is an SPMI
>>> +  controller with wrapping arbitration logic to allow for multiple
>>> +  on-chip devices to control a single SPMI master.
>>> +
>>> +  The PMIC Arbiter can also act as an interrupt controller, providing
>>> +  interrupts to slave devices.
>>> +
>>> +  See Documentation/devicetree/bindings/spmi/spmi.yaml for the generic
>>> +  SPMI controller binding requirements for child nodes.
>>> +
>>> +allOf:
>>> +  - $ref: spmi.yaml#
>>> +
>>> +properties:
>>> +  $nodename:
>>> +    pattern: "^spmi@.*"
>>> +
>>> +  compatible:
>>> +    const: qcom,spmi-pmic-arb
>>> +
>>> +  reg-names:
>>> +    $ref: /schemas/types.yaml#/definitions/string-array
>>
>> reg-names already has a type defined.
> I understand there is a pattern property defined in dt-core.yaml and it 
> defines ".*-names" as a "non-unique-string-array" type. But here, the 
> strings in "reg-names" needs to be unique and it has to be ["core", 
> "intr", "cnfg"] or ["core", "intr", "cnfg", "chnls", "obsrvr"] , that's 
> why I redefined it as "string-array" type which requires each string to 
> be unique. Otherwise, if any dtsi nodes define the "reg-name" as 
> ["core", "core", "core"] will not be caught as a fault.
> 
>>
>>> +    anyOf:
>>> +      - minItems: 3
>>> +      - maxItems: 3
>>> +      - enum: ["core", "intr", "cnfg"]
>>> +
>>> +      - minItems: 5
>>> +      - maxItems: 5
>>> +      - enum: ["core", "intr", "cnfg", "chnls", "obsrvr"]
>>
>> I think you want something like this:
>>
>> minItems: 3
>> items:
>>    - const: core
>>    - const: intr
>>    - const: cnfg
>>    - const: chnls
>>    - const: obsrvr
>>
>>
> As I said, the content for "reg-names" here only has two options , 
> either ["core", "intr", "cnfg"] or ["core", "intr", "cnfg", "chnls", 
> "obsrvr"]. In patch V3, I defined it as below and "make dtbs_check" 
> threw out warnings because some of existing nodes defined "reg-names" 
> with these strings are not having the same order as I defined here (I 
> understood from the warnings that const items need to be followed 
> strictly even in order wise, is this correct?), and I guess the order of 
> the strings doesn't matter here and the schema here shouldn't have such 
> limitation, so I updated it as the "array-string" type and specified the 
> tuples can only be one of the strings defined in the enum. With this, 
> the previous warning regarding "reg-names" in "make dtbs_check" are all 
> fixed.
> 
>    reg-names:
>      oneOf:
>        - items:
>            - const: core
>            - const: intr
>            - const: cnfg
>        - items:
>            - const: core
>            - const: intr
>            - const: cnfg
>            - const: chnls
>            - const: obsrvr
>
Can you help to confirm if I need to change this back to what has been 
defined in PATCH v3 but just ignore those "make dtbs_check" warnings?
Thanks

> 
>>> +
>>> +  reg:
>>> +    minItems: 3
>>> +    maxItems: 5
>>> +    description: |
>>> +      Specifies base physical address and size of the registers in 
>>> SPMI PMIC
>>> +      Arbiter HW module, with the following order.
>>> +        - SPMI PMIC arbiter core registers (core)
>>> +        - SPMI PMIC arbiter interrupt controller registers (intr)
>>> +        - SPMI PMIC arbiter configuration registers (cnfg)
>>> +        - SPMI PMIC arbiter tx-channel per virtual slave registers 
>>> (chnls)
>>> +        - SPMI PMIC arbiter rx-channel per virtual slave registers 
>>> (obsrvr).
>>> +      Register for "chnls" and "obsrvr" are only applicable for PMIC 
>>> arbiter
>>> +      with HW version greater than V2.
>>> +
>>> +  "#address-cells":
>>> +    const: 2
>>> +
>>> +  "#size-cells":
>>> +    const: 0
>>> +
>>> +  interrupts:
>>> +    description: The summary interrupt for the PMIC Arb controller.
>>> +    maxItems: 1
>>> +
>>> +  interrupt-names:
>>> +    const: periph_irq
>>> +
>>> +  interrupt-controller: true
>>> +
>>> +  "#interrupt-cells":
>>> +    const: 4
>>> +    description: |
>>> +      Specifies the number of cells needed to encode any interrupt 
>>> source.
>>> +      The 1st cell is the slave ID for the requested interrupt, its 
>>> valid
>>> +      range is [0-15].
>>> +      The 2nd cell is the  peripheral ID for requested interrupt, 
>>> its valid
>>> +      range is [0-255].
>>> +      The 3rd cell is the requested peripheral interrupt, its valid 
>>> range
>>> +      is [0-7].
>>> +      The 4th cell is interrupt flags indicating level-sense 
>>> information,
>>> +      as defined in dt-bindings/interrupt-controller/irq.h
>>> +
>>> +  qcom,ee:
>>> +    description: the active Execution Environment identifier
>>> +    $ref: /schemas/types.yaml#/definitions/uint32
>>> +    enum: [0, 1, 2, 3, 4, 5]
>>> +
>>> +  qcom,channel:
>>> +    description: which of the PMIC Arbiter provided channels to use 
>>> for accesses
>>> +    $ref: /schemas/types.yaml#/definitions/uint32
>>> +    enum: [0, 1, 2, 3, 4, 5]
>>> +
>>
>>> +patternProperties:
>>> +  "@[0-9a-f]$":
>>> +    description: up to 16 child PMIC nodes
>>> +    type: object
>>> +
>>> +    properties:
>>> +      reg:
>>> +        items:
>>> +          - minItems: 1
>>> +            items:
>>> +              - minimum: 0
>>> +                maximum: 0xf
>>> +              - enum: [ 0 ]
>>> +                description:
>>> +                  0 means user ID address. 1 is reserved for group ID
>>> +                  address.
>>> +
>>> +    required:
>>> +      - reg
>>
>> All this should be covered by spmi.yaml
>>
>>> +
>>> +required:
>>> +  - compatible
>>> +  - reg-names
>>> +  - reg
>>> +  - "#address-cells"
>>> +  - "#size-cells"
>>> +  - qcom,ee
>>> +  - qcom,channel
>>> +
>>> +additionalProperties: false
>>> +
>>> +examples:
>>> +  - |
>>> +    spmi@fc4cf000 {
>>> +          compatible = "qcom,spmi-pmic-arb";
>>> +          reg-names = "core", "intr", "cnfg";
>>> +          reg = <0xfc4cf000 0x1000>,
>>> +                <0xfc4cb000 0x1000>,
>>> +                <0xfc4ca000 0x1000>;
>>> +          interrupt-names = "periph_irq";
>>> +          interrupts = <0 190 0>;
>>> +          interrupt-controller;
>>> +          #interrupt-cells = <4>;
>>> +
>>> +          qcom,ee = <0>;
>>> +          qcom,channel = <0>;
>>> +
>>> +          #address-cells = <2>;
>>> +          #size-cells = <0>;
>>> +    };
>>> -- 
>>> 2.7.4
>>>
>>>

      reply	other threads:[~2022-01-07  7:03 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-21  7:19 [PATCH v4 00/11] A bunch of fix and optimization patches in spmi-pmic-arb.c Fenglin Wu
2021-12-21  7:19 ` [PATCH v4 01/11] spmi: pmic-arb: add a print in cleanup_irq Fenglin Wu
2021-12-21  7:20 ` [PATCH v4 02/11] spmi: pmic-arb: handle spurious interrupt Fenglin Wu
2021-12-21  7:20 ` [PATCH v4 03/11] spmi: pmic-arb: do not ack and clear peripheral interrupts in cleanup_irq Fenglin Wu
2021-12-21  7:20 ` [PATCH v4 04/11] spmi: pmic-arb: check apid against limits before calling irq handler Fenglin Wu
2021-12-21  7:20 ` [PATCH v4 05/11] spmi: pmic-arb: add support to dispatch interrupt based on IRQ status Fenglin Wu
2021-12-21  7:20 ` [PATCH v4 06/11] spmi: pmic-arb: correct duplicate APID to PPID mapping logic Fenglin Wu
2021-12-21  7:20 ` [PATCH v4 07/11] spmi: pmic-arb: block access for invalid PMIC arbiter v5 SPMI writes Fenglin Wu
2021-12-21  7:20 ` [PATCH v4 08/11] bindings: spmi: spmi-pmic-arb: mark interrupt properties as optional Fenglin Wu
2021-12-21 15:13   ` Rob Herring
2021-12-22  0:48     ` Fenglin Wu
2021-12-21  7:20 ` [PATCH v4 09/11] spmi: pmic-arb: make interrupt support optional Fenglin Wu
2021-12-21  7:20 ` [PATCH v4 10/11] spmi: pmic-arb: increase SPMI transaction timeout delay Fenglin Wu
2021-12-21  7:20 ` [PATCH v4 11/11] dt-bindings: convert qcom,spmi-pmic-arb binding to YAML format Fenglin Wu
2021-12-21 11:11   ` Rob Herring
2021-12-21 23:41     ` Fenglin Wu
2021-12-21 14:42   ` Rob Herring
2021-12-22  0:45     ` Fenglin Wu
2022-01-07  7:03       ` Fenglin Wu [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5ef25022-b53b-4002-7de8-e31bfb236009@quicinc.com \
    --to=quic_fenglinw@quicinc.com \
    --cc=agross@kernel.org \
    --cc=bjorn.andersson@linaro.org \
    --cc=collinsd@codeaurora.org \
    --cc=devicetree@vger.kernel.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maz@kernel.org \
    --cc=quic_subbaram@quicinc.com \
    --cc=robh@kernel.org \
    --cc=sboyd@kernel.org \
    --cc=subbaram@codeaurora.org \
    --cc=tglx@linutronix.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®