Re: [PATCH 01/15] soc: qcom: geni-se: allow selection on 32-bit ARM Qualcomm SoCs

On 8/11/2026 3:14 PM, Krzysztof Kozlowski wrote:
> On 11/08/2026 03:15, Dmitry Baryshkov wrote:
>> On Tue, Aug 11, 2026 at 03:48:00AM +0300, Dmitry Baryshkov wrote:
>>> On Sun, Aug 09, 2026 at 12:57:00PM +0530, Hardeep Sharma wrote:
>>>> Commit e10b198a94c6 ("soc: qcom: Restrict drivers per ARM/ARM64")
>>>> tightened QCOM_GENI_SE's dependency to ARM64 || COMPILE_TEST on the
>>>> assumption that no ARMv7 Qualcomm SoC uses the GENI Serial Engine
>>>> hardware.
>>>>
>>>> Kuno, a 32-bit ARM Cortex-A7 Qualcomm SoC, uses the GENI-based QUP
>>>> wrapper for its debug UART, so this assumption no longer holds.
>>>>
>>>> Fixes: e10b198a94c6 ("soc: qcom: Restrict drivers per ARM/ARM64")
> I doubt this fixes tag, otherwise please explain the bug.
>
> How Kuno could be supported, as you claim in e10b198a94c6, if we do not
> support non-dts platforms and there is no Kuno DTS?
>
>>>> Signed-off-by: Hardeep Sharma <[email protected]>
>>>> ---
>>>>   drivers/soc/qcom/Kconfig | 1 -
>>>>   1 file changed, 1 deletion(-)
>>>>
>>>> diff --git a/drivers/soc/qcom/Kconfig b/drivers/soc/qcom/Kconfig
>>>> index fd4d4ecd2df0..5b3c9d31fd7f 100644
>>>> --- a/drivers/soc/qcom/Kconfig
>>>> +++ b/drivers/soc/qcom/Kconfig
>>>> @@ -39,7 +39,6 @@ config QCOM_COMMAND_DB
>>>>   
>>>>   config QCOM_GENI_SE
>>>>   	tristate "Qualcomm GENI Serial Engine Driver"
>>>> -	depends on ARM64 || COMPILE_TEST
>>> depends on ARM || ARM64 || COMPILE_TEST
>> Or depends on ARCH_QCOM || COMPILE_TEST
> None of them are needed, because this is part of qcom soc.
>
> Best regards,
> Krzysztof
Thanks Krzysztof for review.

I'll fix this in v2:
     - drop the Fixes: e10b198a94c6 tag;
     - reword the commit message from "fix" to enablement framing;

Regards

Hardeep Sharma