On 20/01/2026 11:15, Borislav Petkov wrote:

On Thu, Jan 15, 2026 at 12:17:17PM +0000, Ahmed Tiba wrote:

quoted

In drivers/acpi/apei/ghes.c, this subset of CPER handling
is already implemented using a number of helpers mostly named
estatus_* rather than cper_*.

Which helpers are those?

$ git grep -Ei "\Westatus_" drivers/acpi/apei/ghes.c

doesn't give me a single one. Maybe I can't grep.

A simpler grep that catches both prefixes and suffixes is:

git grep -E "estatus_|_estatus" drivers/acpi/apei/ghes.c

quoted

The naming therefore originates from the existing GHES implementation, not
from a new abstraction.

And even if, calling it a "estatus core" and making it a separate thing is
just confusing and unnecessary.

Agreed. I’ll drop the “core” framing and keep it as a plain shared 
compilation unit with exported helpers.

quoted

What I did was lift that existing estatus_* logic so it can be reused by a
non-ACPI provider, rather than duplicating the CPER handling
in a parallel DT path.

Nothing wrong with that as long as you simply export common helpers and we
agree on how they're going to be shared. We do stuff like that all the time.

If you have a reference series that demonstrates the preferred pattern 
for doing this, I’d appreciate a pointer and will follow that approach.

quoted

Because the DT-based path still needs a clean way to call into the shared
logic without dragging in ACPI plumbing.

Yes, carve them out in a separate compilation unit and select it in your DT
glue and that's it. No need for a Kconfig item and calling it a "core" bla.

Works for me. The only question I had is where you’d prefer that unit to 
live so it can be built without pulling in ACPI/APEI (not gated by 
CONFIG_ACPI), while still staying under the ACPI/APEI maintainer umbrella.

Would you prefer to:

1- Keep the shared unit under drivers/acpi/apei/ but make it buildable
    independent of CONFIG_ACPI, and have both GHES + DT select it.

2- Place the shared unit somewhere APEI-adjacent but not ACPI-gated so
    it remains reviewed by the same folks, and have GHES/DT both link it.

Thanks,
Ahmed