fel: add Allwinner A523 SoC support #202
Conversation
The Allwinner A523 has 128KiB of MCU0 SRAM, which the BootROM will not touch. The BROM will use some memory in SRAM A2, which it also clears upon entering FEL mode, starting from address 0x44000. The lowest allocation seems to be the IRQ stack growing down from 0x45400. So we won't touch any of that memory, but can freely use the full 128KB of the primary SRAM for payloads. This means we won't need to swap any buffers for preserving BROM stacks. We put the SPL thunk code just below 0x44000, to leave as much SRAM for the payload as possible. The rest of the SoC is pretty standard, although the watchdogs are now in separate MMIO frames, not part of some timer block anymore. The secure boot mode will prevent even reading the BootROM, so we can use an address in there to test for the secure boot state. However, even though a simple "smc #0" will return to its caller, the NS bit is still set, so we are still in non-secure state afterwards. So leave this bit out for now until we figure out how to switch to secure state properly. Signed-off-by: Andre Przywara <osp@andrep.de>
|
Great, thanks for looking into it! Seems like an interesting new platform (finally). Have you seen some actual board with it (not a tablet)? |
|
Hi @paulkocialkowski many thanks for having a look and merging it! |
|
Very nice find! I see you are spending much more time than I am looking for new Allwinner hardware. I've seen that aodzip has done some work on A133 DRAM init so I'm considering getting one of those (no community hardware is available though). And lack of HDMI is a major drawback for general-purpose use cases too. |
|
I have an A133 tablet, and have opened it to get access to the UART pins and a FEL pin. And I get it into FEL, and can run code and see output on the UART. However it's stuck in non-secure state as well, even though I can access SRAM A2, so I doubt it's using secure boot. I cannot boot from SD card though, so I am a bit stuck on there. It's a lower hanging fruit, upstream-wise, since the basics are in for a while now (the A100 is reportedly just a lower bin of the A133), and I have already rebased the second bunch of patches that were already on the list back then. But without secure state to reset into AArch64 it's a bit of a lost cause :-( |
|
oh, forgot:
Well, I extracted the boot0 from a recovery image, so I guess it's just a question of diligence and persistence to get this reverse engineered (or get @jernejsk interested ;-). In the past I managed to hack the vendor boot0 to return to FEL after DRAM init, so this code can be used to unblock further development, and get U-Boot proper and Linux to boot. |
The Allwinner A523 has 128KiB of MCU0 SRAM, which the BootROM will not touch. The BROM will use some memory in SRAM A2, which it also clears upon entering FEL mode, starting from address 0x44000. The lowest allocation seems to be the IRQ stack growing down from 0x45400. So we won't touch any of that memory, but can freely use the full 128KB of the primary SRAM for payloads. This means we won't need to swap any buffers for preserving BROM stacks.
We put the SPL thunk code just below 0x44000, to leave as much SRAM for the payload as possible.
The rest of the SoC is pretty standard, although the watchdogs are now in separate MMIO frames, not part of some timer block anymore.
The secure boot mode will prevent even reading the BootROM, so we can use an address in there to test for the secure boot state. However, even though a simple "smc #0" will return to its caller, the NS bit is still set, so we are still in non-secure state afterwards. So leave this bit out for now until we figure out how to switch to secure state properly.