[PATCH 28/34] sunxi: armv8: FEL: save and restore SP_IRQ

[Andre Przywara]

On Sun, 23 Mar 2025 13:26:34 +0100
Jernej Škrabec <jernej.skrabec at gmail.com> wrote:

> Dne nedelja, 23. marec 2025 ob 12:35:38 Srednjeevropski standardni čas je Andre Przywara napisal(a):
> > Thanks for Jernej's JTAG debugging effort, it turns out that the BROM
> > expects SP_IRQ to be saved and restored, when we want to enter back into
> > FEL after the SPL's AArch64 stint.
> > Save and restore SP_IRQ as part of the FEL state handling. The banked
> > MRS/MSR access to SP_IRQ, without actually being in IRQ mode, was
> > introduced with the ARMv7 virtualisation extensions. The Arm Cortex-A8
> > cores used in the A10/A13s or older F1C100s SoCs would not support that,
> > but this code here is purely in the ARMv8/AArch64 code path, so it's
> > safe to use unconditionally.
> > 
> > Reported-by: Jernej Skrabec <jernej.skrabec at gmail.com>
> > Signed-off-by: Andre Przywara <andre.przywara at arm.com>  
> 
> I have sneaky suspicion that this is already the issue on H616, but I
> haven't yet confirmed.

That's actually a very good point: if we reset the core, SP_IRQ should
be reset as well - on all 64-bit SoCs, so also A64, H5, H6 and H616. We
are surely not in an IRQ handler when entering the SPL, so the stack
*content* is not relevant, but it should still point to some writable
memory. So I keep scratching my head how this worked so far. If I find
some time, I will try to dump the SP_IRQ content after reset, on those
SoCs.

> FWIW:
> Reviewed-by: Jernej Skrabec <jernej.skrabec at gmail.com>

Thanks for that!

Cheers,
Andre

> 
> Best regards,
> Jernej
> 
> 
>