[RFC 1/4] ARM: tegra: Move SoC drivers to drivers/soc/tegra

[Peter De Schrijver]

> > 
> > I think you should keep the Hyp support in mind anyway and test Linux
> > booting in NS mode on your platforms. It may not be on your radar today
> > for a SoC but if you later decide that it is, we end up with needing
> > different code paths in Linux for the same SoC depending on how it boots
> > (and you can't easily detect this at boot time, nor you can pass it via
> > DT).
> 
> You can just look for a psci DT node? If this node is missing, fallback to
> the existing tegra specific mechanism. At least, that's how I envision doing
> PSCI for HYP mode on v7 based boards where linux today runs in secure mode.
> u-boot loads a minimal secure monitor which only does PSCI (we don't care about
> any security as there are no keys available for signing on those boards
> anyway), adds the PSCI node to the DT and then boots linux in HYP mode. 
> tegra_cpu_reset_handler_enable() checks for a PSCI node and uses it if
> available. Else it will fallback to the existing methods. Similar for cpuidle.
> No guarantees that this will ever be implemented though :) Maybe.
> 

Come to think of it, we follow almost the same approach for Tegra114 where we
support boards with Trusted Foundations secure monitor and linux in secure
mode with the same kernel image.

Cheers,

Peter.