Re: [Patch 1/2] UefiCpuPkg: Remove FIT based microcode shadow logic from MpInitLib.

["Ni, Ray" <ray.ni@intel.com>]

By the way, please rebase to the latest code when sending out
the V2 patch.

> -----Original Message-----
> From: Ni, Ray
> Sent: Tuesday, February 11, 2020 7:21 PM
> To: Fu, Siyuan <siyuan.fu@intel.com>; devel@edk2.groups.io
> Cc: Dong, Eric <eric.dong@intel.com>; Laszlo Ersek <lersek@redhat.com>; Kinney, Michael D <michael.d.kinney@intel.com>
> Subject: RE: [Patch 1/2] UefiCpuPkg: Remove FIT based microcode shadow logic from MpInitLib.
> 
> 
> > +typedef
> > +EFI_STATUS
> > +(EFIAPI *EDKII_PEI_SHADOW_MICROCODE) (
> > +  IN  EDKII_PEI_SHADOW_MICROCODE_PPI        *This,
> > +  IN  UINTN                                 CpuIdCount,
> > +  IN  EDKII_PEI_CPU_MICROCODE_ID            *MicrocodeCpuId,
> 
> 1. How about CpuMicrocodeId or EDKII_PEI_MICROCODE_CPU_ID?
> I'd like to keep the name and type be matched.
> 
> 
> > +  OUT UINTN                                 *BufferSize,
> > +  OUT VOID                                  **Buffer
> 2. I remember that we offline discussed that Buffer/BufferSize are not needed
> to be part of the parameters. It can provide better flexibility that doesn't require
> the microcode in memory is in continuous memory.
> Why are they still in the parameters?
> 
> OK. I see now. Because EDKII_MICROCODE_PATCH_HOB contains below fields:
> typedef struct {
>   UINT64    MicrocodePatchAddress;
>   UINT64    MicrocodePatchRegionSize;
>   ...
> } EDKII_MICROCODE_PATCH_HOB;
> which already restricts that the microcode in memory is in continuous memory.
> 
> I'm ok with this.
> 
> > +EFI_STATUS
> > +PlatformShadowMicrocode (
> > +  IN OUT CPU_MP_DATA             *CpuMpData
> > +  )
> > +{
> > +  return EFI_NOT_FOUND;
> 
> 3. Can you add comments to say that microcode shadow
> in DXE only supports the location identified by PCD?
> 4. How about returning EFI_UNSUPPORTED?