linux-kernel - Re: [PATCH v2 5/5] arm64: dts: qcom: sm8250: Add GPU speedbin 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 for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <ff4e3f7a-04eb-2049-be49-e193a6ee2b2d@linaro.org>
Date:   Fri, 31 Mar 2023 03:59:32 +0300
From:   Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
To:     Konrad Dybcio <konrad.dybcio@...aro.org>,
        linux-arm-msm@...r.kernel.org, andersson@...nel.org,
        agross@...nel.org, krzysztof.kozlowski@...aro.org
Cc:     marijn.suijten@...ainline.org, Rob Herring <robh+dt@...nel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
        devicetree@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 5/5] arm64: dts: qcom: sm8250: Add GPU speedbin support

On 31/03/2023 03:43, Dmitry Baryshkov wrote:
> On 20/01/2023 19:22, Konrad Dybcio wrote:
>> SM8250 has (at least) four GPU speed bins. With the support added on the
>> driver side, wire up bin detection in the DTS to restrict lower-quality
>> SKUs from running at frequencies they were not validated at.
>>
>> Tested-by: Marijn Suijten <marijn.suijten@...ainline.org> # On Sony 
>> Xperia 5 II (speed bin 0x7)
>> Reviewed-by: Marijn Suijten <marijn.suijten@...ainline.org>
>> Signed-off-by: Konrad Dybcio <konrad.dybcio@...aro.org>
>> ---
>>   arch/arm64/boot/dts/qcom/sm8250.dtsi | 23 ++++++++++++++++++++++-
>>   1 file changed, 22 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/arm64/boot/dts/qcom/sm8250.dtsi 
>> b/arch/arm64/boot/dts/qcom/sm8250.dtsi
>> index 059c83003fb6..95f1a6afcd43 100644
>> --- a/arch/arm64/boot/dts/qcom/sm8250.dtsi
>> +++ b/arch/arm64/boot/dts/qcom/sm8250.dtsi
>> @@ -981,6 +981,18 @@ ipcc: mailbox@...000 {
>>               #mbox-cells = <2>;
>>           };
>> +        qfprom: efuse@...000 {
>> +            compatible = "qcom,sm8250-qfprom", "qcom,qfprom";
>> +            reg = <0 0x00784000 0 0x8ff>;
>> +            #address-cells = <1>;
>> +            #size-cells = <1>;
>> +
>> +            gpu_speed_bin: gpu_speed_bin@19b {
>> +                reg = <0x19b 0x1>;
>> +                bits = <5 3>;
>> +            };
>> +        };
>> +
>>           rng: rng@...000 {
>>               compatible = "qcom,prng-ee";
>>               reg = <0 0x00793000 0 0x1000>;
>> @@ -2576,49 +2588,58 @@ gpu: gpu@...0000 {
>>               qcom,gmu = <&gmu>;
>> +            nvmem-cells = <&gpu_speed_bin>;
>> +            nvmem-cell-names = "speed_bin";
>> +
>>               status = "disabled";
>>               zap-shader {
>>                   memory-region = <&gpu_mem>;
>>               };
>> -            /* note: downstream checks gpu binning for 670 Mhz */
>>               gpu_opp_table: opp-table {
>>                   compatible = "operating-points-v2";
>>                   opp-670000000 {
>>                       opp-hz = /bits/ 64 <670000000>;
>>                       opp-level = <RPMH_REGULATOR_LEVEL_NOM_L1>;
>> +                    opp-supported-hw = <0x6>;
> 
> opp-supported-hw = <0xa>; /* 3 & 1 */

Konrad pointed out that vendor dtsi has pwrlevel #3 for speed-bin=2 and 
vice versa, so the patch is correct (it has mapped speed-bin to hw as 
1:1 instead of weird 2<->3 mapping).

Ideally there should be a comment regarding this.
Nevertheless:

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>

> 
>>                   };
>>                   opp-587000000 {
>>                       opp-hz = /bits/ 64 <587000000>;
>>                       opp-level = <RPMH_REGULATOR_LEVEL_NOM>;
>> +                    opp-supported-hw = <0x7>;
> 
> opp-supported-hw = <0xb>; /* 3, 1, 0 */
> 
> 
>>                   };
>>                   opp-525000000 {
>>                       opp-hz = /bits/ 64 <525000000>;
>>                       opp-level = <RPMH_REGULATOR_LEVEL_SVS_L2>;
>> +                    opp-supported-hw = <0xf>;
>>                   };
>>                   opp-490000000 {
>>                       opp-hz = /bits/ 64 <490000000>;
>>                       opp-level = <RPMH_REGULATOR_LEVEL_SVS_L1>;
>> +                    opp-supported-hw = <0xf>;
>>                   };
>>                   opp-441600000 {
>>                       opp-hz = /bits/ 64 <441600000>;
>>                       opp-level = <RPMH_REGULATOR_LEVEL_SVS_L0>;
>> +                    opp-supported-hw = <0xf>;
>>                   };
>>                   opp-400000000 {
>>                       opp-hz = /bits/ 64 <400000000>;
>>                       opp-level = <RPMH_REGULATOR_LEVEL_SVS>;
>> +                    opp-supported-hw = <0xf>;
>>                   };
>>                   opp-305000000 {
>>                       opp-hz = /bits/ 64 <305000000>;
>>                       opp-level = <RPMH_REGULATOR_LEVEL_LOW_SVS>;
>> +                    opp-supported-hw = <0xf>;
>>                   };
>>               };
>>           };
> 

-- 
With best wishes
Dmitry