Re: [PATCH 1/2] armv8: aarch32: Execute 32-bit Linux for LayerScape platforms

From: Robin Murphy
Date: Fri Sep 23 2016 - 11:13:39 EST


On 23/09/16 15:44, Arnd Bergmann wrote:
> On Friday, September 23, 2016 3:24:12 PM CEST Robin Murphy wrote:
>> On 23/09/16 15:01, Stuart Yoder wrote:
>> Otherwise you can
>> always simply run your own shim at EL2 to drive an AArch32 EL1 (it'll
>> need to trap and translate subsequent SMC calls for e.g. PSCI).
>>
>>> If there is such a requirement, it's something begging for standardization.
>>> Doesn't make sense for multiple divergent approaches for switching from
>>> aarch64/EL2 to aarch32/EL2.
>>
>> Perhaps - I did briefly look into how hard it would be to write a proper
>> SMC service handler to do this (since ATF does have a framework for such
>> things), but concluded it would be more than 10 minutes' work and just
>> cheated instead. It's certainly something which could be raised with the
>> firmware folks.
>
> If we end up allowing all arm64 platforms to be enabled in arch/arm,
> we could perhaps create a generic implementation that does both of
> those things, i.e.
>
> - Take the arm32 kernel Image or zImage file, wrap it inside of a binary
> that implements the arm64 boot protocol.
> - When that starts up, try to use the new PSCI call to jump into
> the arm32 kernel
> - If PSCI failed and we are running in EL2, implement the shim
> and start the arm32 kernel in EL1 mode

Really, though, the firmware call thing is an incredibly niche use-case.
Beyond development, the only real benefit of starting an AArch32 kernel
in Hyp is that you can run AArch32 KVM guests, which you can do equally
well (if not better) under an AArch64 kernel.

The standalone EL2 shim probably has legs, though, if the one currently
found in arch/arm64 is too heavyweight ;) My first version was
essentially that, albeit implemented by a load of .inst directives in
head.S.

Robin.

>
> Arnd
>