linux-kernel - Re: [PATCH 1/2] arch: arm64: qcom: sdm845-shift-axolotl: fix Bluetooth firmware loading

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: <0488a37f-083c-6e8c-eff2-e204b242cbfb@connolly.tech>
Date:   Wed, 06 Jul 2022 18:49:44 +0000
From:   Caleb Connolly <caleb@...nolly.tech>
To:     Dmitry Baryshkov <dmitry.baryshkov@...aro.org>,
        Alexander Martinz <amartinz@...ftphones.com>
Cc:     ~postmarketos/upstreaming@...ts.sr.ht, phone-devel@...r.kernel.org,
        Andy Gross <agross@...nel.org>,
        Bjorn Andersson <bjorn.andersson@...aro.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,
        Dylan Van Assche <me@...anvanassche.be>
Subject: Re: [PATCH 1/2] arch: arm64: qcom: sdm845-shift-axolotl: fix Bluetooth firmware loading



On 09/06/2022 13:47, Dmitry Baryshkov wrote:
> On Thu, 9 Jun 2022 at 12:54, Alexander Martinz <amartinz@...ftphones.com> wrote:
>>
>> From: Dylan Van Assche <me@...anvanassche.be>
>>
>> Add hsuart0 alias, firmware name and prevent garbage data on Bluetooth UART lines
>
> hsuart aliases are not standard. Please use serialN alias instead.
>
>> on the SHIFT 6mq based on the Qualcomm SDM845 chip.
>> I discovered that these were missing by comparing the DTS with similar
>> devices such as the Oneplus 6/6T and Dragonboard 845c.
>>
>> Signed-of-by: Dylan Van Assche <me@...anvanassche.be>
>> Tested-by: Alexander Martinz <amartinz@...ftphones.com>
>> ---
>>   .../boot/dts/qcom/sdm845-shift-axolotl.dts    | 33 +++++++++++++++++++
>>   1 file changed, 33 insertions(+)
>>
>> diff --git a/arch/arm64/boot/dts/qcom/sdm845-shift-axolotl.dts b/arch/arm64/boot/dts/qcom/sdm845-shift-axolotl.dts
>> index 103cc40816fd..fa72f23ef0c2 100644
>> --- a/arch/arm64/boot/dts/qcom/sdm845-shift-axolotl.dts
>> +++ b/arch/arm64/boot/dts/qcom/sdm845-shift-axolotl.dts
>> @@ -20,6 +20,7 @@ / {
>>
>>          aliases {
>>                  display0 = &framebuffer0;
>> +               hsuart0 = &uart6;
>>                  serial0 = &uart9;
>>          };
>>
>> @@ -529,6 +530,32 @@ volume_down_resin: resin {
>>          };
>>   };
>>
>> +/*
>> + * Prevent garbage data on bluetooth UART lines
>> + */
>> +&qup_uart6_default {
>> +        pinmux {
>> +                pins = "gpio45", "gpio46", "gpio47", "gpio48";
>> +                function = "qup6";
>> +        };
>
> This chunk should probably go into sdm845.dtsi. I'd suggest either
> adding a separate qup_uart6_4pin or just making qup_uart6_default the
> 4-pin UART.
This sounds good
>
>> +
>> +        cts {
>> +                pins = "gpio45";
>
> Then the label and 'pins' can go into the dtsi, while keeping just the
> bias and drive-strength in the platform dtsi.
>
>> +                bias-pull-down;
>> +        };
>> +
>> +        rts-tx {
>> +                pins = "gpio46", "gpio47";
>> +                drive-strength = <2>;
>> +                bias-disable;
>> +        };
>> +
>> +        rx {
>> +                pins = "gpio48";
>> +                bias-pull-up;
>> +        };
>> +};
>> +
>>   &qup_uart9_default {
>>          pinconf-rx {
>>                  pins = "gpio5";
>> @@ -667,6 +694,12 @@ &uart6 {
>>          bluetooth {
>>                  compatible = "qcom,wcn3990-bt";
>>
>> +               /*
>> +                * This path is relative to the qca/
>> +                * subdir under lib/firmware.
>> +                */
>> +               firmware-name = "axolotl/crnv21.bin";
>
> Do you know what is the difference between axolotl's bin and the regular one?
Just checked in a hex editor and there are some differences. This firmware is
signed on most devices but given the SHIFT6mq is secureboot off we could in
theory use upstream firmware here - if it works. This particular device has some
very broken wifi firmware - we have to use the firmware from the OnePlus 6 to
get wifi to work at all...
>
>> +
>>                  vddio-supply = <&vreg_s4a_1p8>;
>>                  vddxo-supply = <&vreg_l7a_1p8>;
>>                  vddrf-supply = <&vreg_l17a_1p3>;
>
> --
> With best wishes
> Dmitry

--
Kind Regards,
Caleb