Re: [PATCH v6] i386/cpu: fixup number of addressable IDs for logical processors in the physical package

[Zhao Liu]

On Tue, Dec 03, 2024 at 11:04:12PM +0800, Xiaoyao Li wrote:
> Date: Tue, 3 Dec 2024 23:04:12 +0800
> From: Xiaoyao Li <xiaoyao.li@intel.com>
> Subject: Re: [PATCH v6] i386/cpu: fixup number of addressable IDs for
>  logical processors in the physical package
> 
> On 12/3/2024 3:33 PM, Zhao Liu wrote:
> > > However, back to the patch, I think we cannot change it as this patch
> > > directly. Instead, we need a compat_props for the changed behavior, 
> > > because
> > > this isn't a bug fix and it introduces guest-visible differences.
> > 
> > This is a fix, not a new feature, so compat_props is not needed.
> 
> Fix what? QEMU behaves as it for so many years and if the guest OS uses the
> algorithm recommended by SDM, there is no issue.

I've spent a lot time to explain why current behavior doesn't match the
SDM and real machine's implementation.

> > > For ancient Intel CPUs, EBX[23:16] did represent the number of Logical
> > > processor per package. I believe this should be the reason why QEMU
> > > implemented it as is:
> > > 
> > >    - on SDM version 013, EBX[23:16]: Number of logical processors per
> > > physical processor; two for the Pentium 4 processor supporting
> > > Hyper-Threading Technology.
> > > 
> > >    - on SDM version 015, it changed to: Number of initial APIC IDs 
> > > reserved
> > > for this physical package. Normally, this is the number of logical
> > > processors per physical package.
> > > 
> > >    - on SDM version 016, it changed to: Maximum number of logical 
> > > processors
> > > in this physical package.
> > > 
> > >    - finally, starting from SDM version 026, it changed to what reads now:
> > > Maximum number of addressable IDs for logical processors in this physical
> > > package.
> > 
> > And this is an architecturally defined CPUID, so SDM ensures backward
> > compatibility.
> 
> SDM ensure the backwards compatibility by recommending to round the number
> up to the power-of 2 when using it to calculate the topology with legacy
> method.

Please, *always* refer the latest SDM.

Regarding historical changes, older machines didn't have spare APIC ID
slots, so the actual number is the same as the maximum number of
addressable IDs.