[DEBUG] rk3399-nanopi-neo4: SPL SD init may report -5 due to SD power-on timing

[Sune Brian]

> Can you try to apply
> https://lore.kernel.org/u-boot/20251031145951.535376-1-c.stoidner@phytec.de/
> and see if it helps? It seems like the implem in U-Boot doesn't exactly
> follow the MMC spec and this supposedly fixes it. My RK3399 board has
> the vmmc-supply always on (not configurable by software) so I cannot be
> impacted by any change.
>
> If this isn't enough, you need to specify in the regulator node how long
> it takes for the regulator to be stable. We don't want an average, we
> want the worst case scenario so that all boards work, if you only have
> one NanoPi NEO4, then so be it use the delay necessary for yours, people
> can increase the delay later on so it unbreaks their unit as well. This
> is what Jonas recommended already.

> Quentin

Appreciated the suggestions.
But I had resolved the issue.
Turns out it is wrong on the delay path.
This is just my blind guess.
If you read through the mails, me and Joans.

The inherent solution is to comment out the regulator
in the SD node of the dts in order to sense the SD
card, but this is a very bad method.

The actual issue [1] and I have tried (works).
[1] Revert "power: regulator: Add vin-supply for GPIO and Fixed regulators"
https://patchwork.ozlabs.org/project/uboot/patch/20251101203429.1603832-1-jonas@kwiboo.se/


Bests,
Brian