Message-ID: <25f5a750-b51c-7d7b-0d50-5b2f78de8512@linaro.org>
Date:   Fri, 3 Feb 2023 00:45:49 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.7.0
Subject: Re: [RFT PATCH 05/14] arm64: dts: qcom: sc8280xp: correct TLMM
 gpio-ranges
Content-Language: en-US
To:     Brian Masney <bmasney@redhat.com>,
        Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Cc:     Andy Gross <agross@kernel.org>,
        Bjorn Andersson <andersson@kernel.org>,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        linux-arm-msm@vger.kernel.org, devicetree@vger.kernel.org,
        linux-kernel@vger.kernel.org
References: <20230201155105.282708-1-krzysztof.kozlowski@linaro.org>
 <20230201155105.282708-6-krzysztof.kozlowski@linaro.org>
 <Y9xAEoc0QXe222D0@x1>
From:   Konrad Dybcio <konrad.dybcio@linaro.org>
In-Reply-To: <Y9xAEoc0QXe222D0@x1>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 2.02.2023 23:58, Brian Masney wrote:
> On Wed, Feb 01, 2023 at 04:50:56PM +0100, Krzysztof Kozlowski wrote:
>> Correct the number of GPIOs in TLMM pin controller.
>>
>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
>> ---
>>  arch/arm64/boot/dts/qcom/sc8280xp.dtsi | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/arch/arm64/boot/dts/qcom/sc8280xp.dtsi b/arch/arm64/boot/dts/qcom/sc8280xp.dtsi
>> index fa2d0d7d1367..17e8c26a9ae6 100644
>> --- a/arch/arm64/boot/dts/qcom/sc8280xp.dtsi
>> +++ b/arch/arm64/boot/dts/qcom/sc8280xp.dtsi
>> @@ -3533,7 +3533,7 @@ tlmm: pinctrl@f100000 {
>>  			#gpio-cells = <2>;
>>  			interrupt-controller;
>>  			#interrupt-cells = <2>;
>> -			gpio-ranges = <&tlmm 0 0 230>;
>> +			gpio-ranges = <&tlmm 0 0 228>;
Won't that kill the UFS pins?


>>  		};
> 
> I verified that this count matches what's in downstream.
> 
> Reviewed-by: Brian Masney <bmasney@redhat.com>
> 
> 
> However, I noticed in upstream that we're using this reg property:
> 
>    reg = <0 0x0f100000 0 0x300000>;
> 
> Downstream has a different base address and a wider size. Note: I added
> spaces for easy comparison.
> 
>    reg = <  0x0F000000   0x1000000>;
> 
> I don't have access to the appropriate documents to see which is
> correct. I assume the base address in upstream is at least correct since
> pinctrl is working on this platform.
Downstream offsets things in the driver

https://git.codelinaro.org/clo/la/kernel/msm-5.4/-/blob/LV.AU.1.2.3.r1-03600-gen3meta.0/drivers/pinctrl/qcom/pinctrl-direwolf.c#L20

Notice how UFS/QDSD pins addresses differ by 0x1000... up- and downstream too.
I'd imagine Bjorn/Johan/whoever did that used magic PDFs instead of not-very-
tested downstream sources.

Another note, the downstream driver may be incomplete/wrong, as Linux was
not exactly the main usecase of 8280xp so the testing there was most likely
only basic.

Konrad
> 
> Brian
>