Poc: persistant storage

[dehanj]

Description

Draft PR, WIP. Some parts of this are still to be considered as a proof of concept.
TODO: write a better description

Type of change

Please tick any that are relevant to this PR and remove any that aren't.

• Bugfix (non breaking change which resolve an issue)
• Feature (non breaking change which adds functionality)
• Breaking Change (a change which would cause existing functionality to not work as expected)
• Documentation (a change to documentation)

Submission checklist

• My code follows the style guidelines of this project
• I have performed a self-review of my changes
• I have tested and verified my changes on target
• My changes are well written and CI is passing
• I have squashed my work to relevant commits and rebased on main for linear history
• I have added a "Co-authored-by: x" if several people contributed, either pair programming or by squashing commits from different authors.
• I have updated the documentation where relevant (readme, dev.tillitis.se etc.)
• QEMU is updated to reflect changes

[dehanj]

Comment from mc in previous PR (#273)

I have tested the firmware in a patched qemu running in firmware mode all the time. I tried the management app, installed preloaded app, started it, and watched the partition table while all this was happening. The syscalls seem to work fine.

I was a bit surprised that the preloaded app wasn't started automatically but I understand that part isn't implemented yet. Perhaps it's good to wait, because I think it's still unclear (#214) how we even would signal back to the firmware how not to start the preloaded app. On the other hand, perhaps now is the time to experiment with different solutions?

There are more things that aren't yet implemented. The TODO notes are fine, of course, but they're often just shorthand notes and doesn't explain what more needs to be done. I suggest we track these things in issues or just tasks in an issue, but at least flesh them out a little.

Since the current PoC just reuses the BLAKE2S MMIO to get to the firmware syscall handler it requires support in the hardware to watch PC all the time and raise the privelege when PC hits the blessed address stored in BLAKE2S. I'm worried about if this solution is possible to do in the hardware design without changing PicoRV32. It's certainly not easy to emulate in qemu.

I looked a lot in qemu and did some experiments with the other suggestion in #234, that is to write something to an MMIO to trigger the syscall. That is certainly easier to catch in both hardware and qemu, even if affecting PC directly might still be unsolved.