On 14/09/23 21:34, Nishanth Menon wrote:
> On 14:56-20230914, Nitin Yadav wrote:
>> Hi Krzysztof,
>>
>> On 14/09/23 11:57, Krzysztof Kozlowski wrote:
>>> On 13/09/2023 13:47, Nitin Yadav wrote:
>>>> Add support for AM62Q NAND card: X8 NAND EXPANSION
>>>> BOARD card (PROC143E1) for AM62x LP SK board.
>
> Commit message is all too wrong as well. Sigh.
>
>>>>
>>>> Signed-off-by: Nitin Yadav <n-yadav@xxxxxx>
>>>> ---
>>>> arch/arm64/boot/dts/ti/k3-am62-main.dtsi | 29 ++++++++++++++++++++++++
>>>> arch/arm64/boot/dts/ti/k3-am62.dtsi | 2 ++
>>>> 2 files changed, 31 insertions(+)
>>>>
>>>> diff --git a/arch/arm64/boot/dts/ti/k3-am62-main.dtsi b/arch/arm64/boot/dts/ti/k3-am62-main.dtsi
>>>> index 284b90c94da8..e93e79d8083f 100644
>>>> --- a/arch/arm64/boot/dts/ti/k3-am62-main.dtsi
>>>> +++ b/arch/arm64/boot/dts/ti/k3-am62-main.dtsi
>>>> @@ -955,4 +955,33 @@ mcasp2: audio-controller@2b20000 {
>>>> power-domains = <&k3_pds 192 TI_SCI_PD_EXCLUSIVE>;
>>>> status = "disabled";
>>>> };
>>>> + gpmc0: memory-controller@3b000000 {
>>>> + status = "disabled";
>>>
>>> status is never first in DTSI. Really, where did you see such code?
>> Thank for pointing out, Will send a revised version.
>
> GPMC is not functional without board specific interface configuration
> such as pinmux. this approach, in fact is all over the place now and
> discussed in the mailing list multiple times now.
>
> What is missing here is the documentation of the constraints as to why
> it is set as disabled by default.
gpmc nand is only am62x lp sk in am62x series. it has pinmux conflict
with macsp1, so disabling gpmc & elm by default for other am62 series.
For am62x lpsk in overlay macsp1 is disabled.
>
>
>>>
>>>> + compatible = "ti,am64-gpmc";
>>>> + power-domains = <&k3_pds 80 TI_SCI_PD_EXCLUSIVE>;
>>>
>>> First is compatible, second is reg/reg-names/ranges.
>>>
>>>
>>> Best regards,
>>> Krzysztof
>>>
>>
>> --
>> Regards,
>> Nitin
>
--
Regards,
Nitin
