linux-kernel - Re: [PATCH 1/7] arm64: dts: qcom: sc8280xp: remove GCC from CX power domain

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: <6ece10ff-cc8c-04b2-4afd-43db216227e1@linaro.org>
Date:   Mon, 2 Jan 2023 09:43:26 +0100
From:   Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
To:     Konrad Dybcio <konrad.dybcio@...aro.org>,
        Andy Gross <agross@...nel.org>,
        Bjorn Andersson <andersson@...nel.org>,
        Rob Herring <robh+dt@...nel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
        linux-arm-msm@...r.kernel.org, devicetree@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/7] arm64: dts: qcom: sc8280xp: remove GCC from CX power
 domain

On 31/12/2022 00:41, Konrad Dybcio wrote:
> 
> 
> On 30.12.2022 17:00, Krzysztof Kozlowski wrote:
>> Bindings do not allow power-domain property in GCC clock controller and
>> documentation does not indicate that GCC is part of VDD_CX.
>>
>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
>>
>> ---
>>
>> Maybe the bindings should be fixed? Maybe this was added as workaround?
>> Anyway looking at documentation I do not see such relation, except
>> downstream vdd_cx-supply (which is the same as in other SoCs and we do
>> not represent it in upstream).
> Some clocks scale with _CX, which is annotated on downstream with vdd-levels.
> We take care of that by using opp tables in consumer drivers. Usually if
> power-domains is added to a clock controller, it means that at least one of
> the clocks needs the power domain to be on (which.. should be true for CX
> if the ARM part runs anyway, no?), as for example VDD_MX/VDD_GFX may not be
> on at boot and trying to enable such clocks would result in a big kaboom..
> 
> TL;DR: if nothing exploded, it's fine to remove it

According to Bjorn, we should keep the domain.

Best regards,
Krzysztof