linux-kernel - Re: [PATCH 08/14] dt-bindings: ufs: qcom: Add sm6115 binding

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]

Date:   Mon, 5 Sep 2022 12:03:15 +0200
From:   Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
To:     Iskren Chernev <iskren.chernev@...il.com>,
        Bjorn Andersson <bjorn.andersson@...aro.org>,
        Rob Herring <robh+dt@...nel.org>
Cc:     phone-devel@...r.kernel.org, ~postmarketos/upstreaming@...ts.sr.ht,
        linux-arm-msm@...r.kernel.org, devicetree@...r.kernel.org,
        Andy Gross <agross@...nel.org>,
        Konrad Dybcio <konrad.dybcio@...ainline.org>,
        Alim Akhtar <alim.akhtar@...sung.com>,
        Avri Altman <avri.altman@....com>,
        Bart Van Assche <bvanassche@....org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
        linux-scsi@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 08/14] dt-bindings: ufs: qcom: Add sm6115 binding

On 05/09/2022 09:29, Iskren Chernev wrote:
> 
> 
> On 9/4/22 22:10, Krzysztof Kozlowski wrote:
>> On 03/09/2022 19:54, Iskren Chernev wrote:
>>>
>>>
>>> On 9/1/22 19:11, Krzysztof Kozlowski wrote:
>>>> On 01/09/2022 10:24, Iskren Chernev wrote:
>>>>> Add SM6115 UFS to DT schema.
>>>>>
>>>>> Signed-off-by: Iskren Chernev <iskren.chernev@...il.com>
>>>>> ---
>>>>>  .../devicetree/bindings/ufs/qcom,ufs.yaml     | 26 +++++++++++++++++++
>>>>>  1 file changed, 26 insertions(+)
>>>>>
>>>>> diff --git a/Documentation/devicetree/bindings/ufs/qcom,ufs.yaml b/Documentation/devicetree/bindings/ufs/qcom,ufs.yaml
>>>>> index f2d6298d926c..7c5f6e2e6d4c 100644
>>>>> --- a/Documentation/devicetree/bindings/ufs/qcom,ufs.yaml
>>>>> +++ b/Documentation/devicetree/bindings/ufs/qcom,ufs.yaml
>>>>> @@ -28,6 +28,7 @@ properties:
>>>>>            - qcom,msm8998-ufshc
>>>>>            - qcom,sc8280xp-ufshc
>>>>>            - qcom,sdm845-ufshc
>>>>> +          - qcom,sm6115-ufshc
>>>>>            - qcom,sm6350-ufshc
>>>>>            - qcom,sm8150-ufshc
>>>>>            - qcom,sm8250-ufshc
>>>>> @@ -178,6 +179,31 @@ allOf:
>>>>>            minItems: 1
>>>>>            maxItems: 1
>>>>>
>>>>> +  - if:
>>>>> +      properties:
>>>>> +        compatible:
>>>>> +          contains:
>>>>> +            enum:
>>>>> +              - qcom,sm6115-ufshc
>>>>> +    then:
>>>>> +      properties:
>>>>> +        clocks:
>>>>> +          minItems: 8
>>>>> +          maxItems: 8
>>>>> +        clock-names:
>>>>> +          items:
>>>>> +            - const: core_clk
>>>>> +            - const: bus_aggr_clk
>>>>> +            - const: iface_clk
>>>>> +            - const: core_clk_unipro
>>>>> +            - const: core_clk_ice
>>>>
>>>> Use existing name and put it in the same place as existing variant - sdm845:
>>>> ice_core_clk
>>>
>>> The only problem with sdm845 bindings is the presence of rx_lane1_sync_clk
>>> clock. I'm guessing I could pass zeros there, because it shouldn't be used. Or
>>> it could be moved to last property and then min/maxItems to guard, but that is
>>> a change to something more-or-less immutable.
>>
>> I don't understand - what is the problem here. How presence of some
>> clock affects name of other clock and its place/location in list of clocks?
> 
> qcom,sdm845-ufshc has 9 clocks, one of which is rx_lane1_sync_clk.
> qcom,sm6115-ufshc has 8 clocks (all of the ones in sdm845 without
> rx_lane1_sync_clk). So if I'm understanding correctly, you want to put the
> sm6115 with sdm845, which means re-use the clocks and reg specification from
> sdm845, which means sm6115 will "inherit" this rx_lane1_sync_clk, and then
> I have to put it in DT (otherwise the schema would complain), and I'm asking if
> I can put an empty (i.e <0 0>) value, so schema is satisfied but clock is still
> not really passed.

No. I want to use the same order and naming as sdm845 variant, but in
your own/dedicated if:else: entry. Just do not create inconsistencies
when not needed. If inconsistency is needed here (which I think is not),
please explain more.

Best regards,
Krzysztof