Re: [PATCH] Thermal: intel: hfi: Give HFI instances package scope

[Rafael J. Wysocki]

On Tue, Jul 9, 2024 at 4:13 AM Ricardo Neri
<ricardo.neri-calderon@xxxxxxxxxxxxxxx> wrote:
>
> On Wed, Jul 03, 2024 at 01:33:03PM +0200, Rafael J. Wysocki wrote:
> > On Wed, Jul 3, 2024 at 7:55 AM Zhang Rui <rui.zhang@xxxxxxxxx> wrote:
> > >
> > > The Intel Software Developer's Manual defines the scope of HFI (registers
> > > and memory buffer) as package. Use package scope* in the software
> >
> > "as a package"
> >
> > > representation of an HFI instance.
> > >
> > > Using die scope in HFI instances has the effect of creating multiple,
> > > conflicting, instances for the same package: each instance allocates its
> > > own memory buffer and configures the same package-level registers.
> > > Specifically, only one of the allocated memory buffers can be set in the
> > > MSR_IA32_HW_FEEDBACK_PTR register. CPUs get incorrect HFI data from the
> > > table.
> > >
> > > The problem does not affect current HFI-capable platforms because they
> > > all have single-die processors.
> > >
> > > * We used die scope for HFI instances because there have been processors
> > > in which packages where enumerated as dies. None of those systems support
> >
> > "were"
> >
> > > HFI. If such a system emerged we would need to quirk it.
> > >
> > > Co-developed-by: Chen Yu <yu.c.chen@xxxxxxxxx>
> > > Signed-off-by: Chen Yu <yu.c.chen@xxxxxxxxx>
> > > Signed-off-by: Zhang Rui <rui.zhang@xxxxxxxxx>
> >
> > Ricardo, any concerns?
>
> No concerns. IMO, it is important to document why we were using die scope
> before. Rui has done it.
>
> This patch looks good to me.
>
> FWIW, Reviewed-by: Ricardo Neri <ricardo.neri-calderon@xxxxxxxxxxxxxxx>

Applied as 6.11 material, thanks!