Re: [PATCH v7 2/4] arm64: defconfig: Enable Glymur configs for boot to shell

[Pankaj Patil <pankaj.patil@oss.qualcomm.com>]

On 2/5/2026 9:46 PM, Bjorn Andersson wrote:
> On Thu, Feb 05, 2026 at 05:50:42PM +0530, Pankaj Patil wrote:
>> The serial engine must be properly setup before kernel reaches
>> "init", so UART driver and its dependencies needs to be built in.
>> Enable its dependency clocks,interconnect and pinctrl as built-in
>> to boot Qualcomm's dev platform - Glymur CRD board to UART console with
>> rootfs on nvme storage.
>> DISPCC enabled as module, used for display.
>>
> 
> I'm sorry I missed this earlier, but what are you actually saying in
> this commit message?
> 
> Subject is lacking "Qualcomm" and "boot to shell" doesn't mean anything.
> 
> You're talking about serial engine and init, and how that relates to
> UART needs to be built in. So this must be the most important part of
> your patch? No, those drivers where enabled in November 2018...
> 
> Then follows a problem statement (problem statements are good!), saying
> that we're enabling the dependencies of the UART. This is good as it
> allow us to reach "init" with a working /dev/console (the reason for
> =y).
> 
> But your mixing this up with "rootfs on NVMe storage", which might
> follow from being able to reach "init" with working UART, but it's not
> the same "problem". What I do believe relates to "rootfs on NVMe
> storage" is the inclusion of GLYMUR_TCSRCC=m though?
> 
> Surely there must be a better way to write "Qualcomm's dev platform -
> Glymur CRD board"?
> 
> DISPCC is used for display, but should your statement be read to imply
> that we now have all drivers enabled for "boot to shell" on the display?
> Or did you include it just because it was adjacent?
> 
> 
> The content of the patch looks good, but the commit message is
> incoherent and lacks answers to someone who doesn't already know why the
> options are enabled in the way they are.
> 
> 
> For comparison look at Jingyi's patch here:
> https://lore.kernel.org/all/20251215-knp-dts-v4-2-1541bebeb89f@oss.qualcomm.com/
> 
> Regards,
> Bjorn
> 

Sure, I'll update the commit subject and text to:

arm64: defconfig: Enable configs for Qualcomm Glymur SoC

Enable pinctrl, clocks and interconnect drivers as built-in
in order for serial console to be available before kernel 
reaches "init" on Qualcomm Glymur CRD.
Additionally, booting rootfs from NVMe requires TCSRCC to
be enabled as module
Enable dispcc as module which is a dependency for display enablement

>> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
>> Signed-off-by: Pankaj Patil <pankaj.patil@oss.qualcomm.com>
>> ---
>>  arch/arm64/configs/defconfig | 5 +++++
>>  1 file changed, 5 insertions(+)
>>
>> diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig
>> index b67d5b1fc45b..d3ed35d3e6a4 100644
>> --- a/arch/arm64/configs/defconfig
>> +++ b/arch/arm64/configs/defconfig
>> @@ -647,6 +647,7 @@ CONFIG_PINCTRL_IMX91=y
>>  CONFIG_PINCTRL_IMX93=y
>>  CONFIG_PINCTRL_IMX_SCMI=y
>>  CONFIG_PINCTRL_MSM=y
>> +CONFIG_PINCTRL_GLYMUR=y
>>  CONFIG_PINCTRL_IPQ5018=y
>>  CONFIG_PINCTRL_IPQ5332=y
>>  CONFIG_PINCTRL_IPQ5424=y
>> @@ -1454,6 +1455,9 @@ CONFIG_COMMON_CLK_MT8192_SCP_ADSP=y
>>  CONFIG_COMMON_CLK_MT8192_VDECSYS=y
>>  CONFIG_COMMON_CLK_MT8192_VENCSYS=y
>>  CONFIG_COMMON_CLK_QCOM=y
>> +CONFIG_CLK_GLYMUR_DISPCC=m
>> +CONFIG_CLK_GLYMUR_GCC=y
>> +CONFIG_CLK_GLYMUR_TCSRCC=m
>>  CONFIG_CLK_KAANAPALI_GCC=y
>>  CONFIG_CLK_KAANAPALI_TCSRCC=m
>>  CONFIG_CLK_X1E80100_CAMCC=m
>> @@ -1846,6 +1850,7 @@ CONFIG_INTERCONNECT_IMX8MN=m
>>  CONFIG_INTERCONNECT_IMX8MQ=m
>>  CONFIG_INTERCONNECT_IMX8MP=y
>>  CONFIG_INTERCONNECT_QCOM=y
>> +CONFIG_INTERCONNECT_QCOM_GLYMUR=y
>>  CONFIG_INTERCONNECT_QCOM_KAANAPALI=y
>>  CONFIG_INTERCONNECT_QCOM_MSM8916=m
>>  CONFIG_INTERCONNECT_QCOM_MSM8953=y
>>
>> -- 
>> 2.34.1
>>