Re: [PATCH v0] RISC-V: Use Zkr to seed KASLR base address

[Clément Léger]

On 07/06/2024 20:51, Deepak Gupta wrote:
> On Mon, Jun 03, 2024 at 01:47:40PM +0100, Conor Dooley wrote:
>> On Mon, Jun 03, 2024 at 11:14:49AM +0200, Alexandre Ghiti wrote:
>>> Hi Conor,
>>>
>>> On 31/05/2024 19:31, Conor Dooley wrote:
>>> > On Fri, May 31, 2024 at 12:23:27PM -0400, Jesse Taube wrote:
>>> > > Dectect the Zkr extension and use it to seed the kernel base
>>> address.
>>> > >
>>> > > Detection of the extension can not be done in the typical
>>> fashion, as
>>> > > this is very early in the boot process. Instead, add a trap handler
>>> > > and run it to see if the extension is present.
>>> > You can't rely on the lack of a trap meaning that Zkr is present
>>> unless
>>> > you know that the platform implements Ssstrict. The CSR with that
>>> number
>>> > could do anything if not Ssstrict compliant, so this approach gets a
>>> > nak from me. Unfortunately, Ssstrict doesn't provide a way to detect
>>> > it, so you're stuck with getting that information from firmware.
>>>
>>>
>>> FYI, this patch is my idea, so I'm the one to blame here :)
>>>
>>>
>>> >
>>> > For DT systems, you can actually parse the DT in the pi, we do it
>>> to get
>>> > the kaslr seed if present, so you can actually check for Zkr. With
>>> ACPI
>>> > I have no idea how you can get that information, I amn't an ACPI-ist.
>>>
>>>
>>> I took a look at how to access ACPI tables this early when
>>> implementing the
>>> Zabha/Zacas patches, but it seems not possible.
>>>
>>> But I'll look into this more, this is not the first time we need the
>>> extensions list very early and since we have no way to detect the
>>> presence
>>> of an extension at runtime, something needs to be done.
>>
>> Aye, having remembered that reading CSR_SEED could have side-effects on a
>> system with non-conforming extensions, it'd be good to see if we can
>> actually do this via detection on ACPI - especially for some other
>> extensions that we may need to turn on very early (I forget which ones we
>> talked about this before for). I didn't arm64 do anything with ACPI in
>> the
>> pi code, is the code arch/x86/boot/compressed run at an equivilent-ish
>> point
>> in boot?
> 
> cc: +Clement and Atish
> 
> I don't know all the details but on first glance it seems like instead
> of ACPI,
> may be FWFT is a better place for discovery ?
> https://lists.riscv.org/g/tech-prs/topic/patch_v12_add_firmware/106479571

IMHO, doing discovery in FWFT is not the goal of this extension. I think
the "real" solution would be to wait for the unified discovery task
group to come up with something for that (which is their goal I think) [1]

Clément,

Link: https://github.com/riscv/configuration-structure [1]

> 
> Supervisor could query if Sstrict is implemented and then it can check for
> lack of trap on CSR_SEED or straight-away check for presence of Zkr.
> 
>