On 08.09.21 07:48, Nishanth Menon wrote:
> On 07:25-20210908, Jan Kiszka wrote:
>> On 08.09.21 04:44, Nishanth Menon wrote:
>>> On 18:37-20210907, Jan Kiszka wrote:
>>>> From: Jan Kiszka <jan.kiszka@xxxxxxxxxxx>
>>>>
>>>> This ensures that the SD card will remain mmc0 across devices with and
>>>> without eMMC.
>>>>
>>>> Signed-off-by: Jan Kiszka <jan.kiszka@xxxxxxxxxxx>
>>>> ---
>>>> arch/arm64/boot/dts/ti/k3-am65.dtsi | 2 ++
>>>> 1 file changed, 2 insertions(+)
>>>>
>>>> diff --git a/arch/arm64/boot/dts/ti/k3-am65.dtsi b/arch/arm64/boot/dts/ti/k3-am65.dtsi
>>>> index a9fc1af03f27..785d931a2dd9 100644
>>>> --- a/arch/arm64/boot/dts/ti/k3-am65.dtsi
>>>> +++ b/arch/arm64/boot/dts/ti/k3-am65.dtsi
>>>> @@ -31,6 +31,8 @@ aliases {
>>>> i2c4 = &main_i2c2;
>>>> i2c5 = &main_i2c3;
>>>> ethernet0 = &cpsw_port1;
>>>> + mmc0 = &sdhci1;
>>>> + mmc1 = &sdhci0;
>>>
>>> Jan,
>>>
>>> Responding in context of [1]. Suggestion from Aswath is to do the
>>> following instead at SoC level:
>>> + mmc0 = &sdhci0;
>>> + mmc1 = &sdhci1;
>>>
>>
>> Then I will leave this up to you and will go back to v1 of this patch
>> (as we require it the other way around).
>
> Could you elaborate on the rationale as to why the other way around?
> That would probably provide the reasoning to override the SoC generic
> definitions in the board dts/dtsi as the exception case.
>
As I wrote in v1: We have deployed devices that use mmcblk0 == SD card
in the field. Those should now be migrated to upstream kernel/DT, but
ideally without glitches in the interface.
Jan
--
Siemens AG, T RDA IOT
Corporate Competence Center Embedded Linux
