[PATCH v2 06/50] drm/msm/dpu: correct sm8550 scaler

Dmitry Baryshkov dmitry.baryshkov at linaro.org
Sat Feb 25 23:06:28 UTC 2023


On 24/02/2023 22:51, Abhinav Kumar wrote:
> 
> 
> On 2/13/2023 9:36 AM, neil.armstrong at linaro.org wrote:
>> On 13/02/2023 12:16, Dmitry Baryshkov wrote:
>>> On 13/02/2023 12:41, Neil Armstrong wrote:
>>>> On 12/02/2023 00:12, Dmitry Baryshkov wrote:
>>>>> QSEED4 is a newer variant of QSEED3LITE, which should be used on
>>>>> sm8550. Fix the DPU caps structure and used feature masks.
>>>>
>>>> I found nowhere SM8550 uses Qseed4, on downstream DT, it's written:
>>>>          qcom,sde-qseed-sw-lib-rev = "qseedv3lite";
>>>>          qcom,sde-qseed-scalar-version = <0x3002>;
>>>
>>> And then the techpack tells us starting from 0x3000 the v3lite is v4:
>>>
>>> https://git.codelinaro.org/clo/la/platform/vendor/opensource/display-drivers/-/blob/display-kernel.lnx.5.10.r8-rel/msm/sde/sde_hw_util.c#L59
>>>
>>> https://git.codelinaro.org/clo/la/platform/vendor/opensource/display-drivers/-/blob/display-kernel.lnx.5.10.r8-rel/msm/sde/sde_hw_util.c#L102
>>
>> OK then:
>>
>> Reviewed-by: Neil Armstrong <neil.armstrong at linaro.org>
>>
>>>
> 
> This little bit of confusion is because with downstream, the qseed is a 
> separate usermode library having its own revision. So the SW lib version 
> in this case is not exactly correlating with the scalar HW revision.

Can you possibly spend some more words here? I see that sde_hw_utils.c 
programs scalers slightly different depending on the version of the 
scaler. At some point the SDE driver was reading the register to 
determine the revision. Then it switched to the revision specified in 
the DTS (which, as far as I understand, corresponds to the HW register 
contents).

So, where does SW revision come into the play? (and which library are we 
talking about?). Is the 'v3lite' an SW revision? Or is the 0x3002 an SW 
revision?

> 
> Since upstream DPU only cares about the HW revision of the scaler, we 
> should be going off the qcom,sde-qseed-scalar-version.
> 
> This change LGTM,
> 
> Reviewed-by: Abhinav Kumar <quic_abhinavk at quicinc.com>


-- 
With best wishes
Dmitry



More information about the dri-devel mailing list