Re: [PATCH] x86/mce/amd, EDAC/mce_amd: Add new SMCA bank types

[Yazen Ghannam <yazen.ghannam@amd.com>]

On Sat, Feb 28, 2026 at 04:04:47PM +0100, Borislav Petkov wrote:
> On Mon, Feb 02, 2026 at 05:21:58PM +0000, Yazen Ghannam wrote:
> > Recognize new SMCA bank types and include their short names for sysfs
> > and long names for decoding.
> > 
> > Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
> > ---
> >  arch/x86/include/asm/mce.h    | 11 +++++++++++
> >  arch/x86/kernel/cpu/mce/amd.c | 21 +++++++++++++++++++++
> >  drivers/edac/mce_amd.c        | 10 ++++++++++
> >  3 files changed, 42 insertions(+)
> > 
> > diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h
> > index 2d98886de09a..6e1f10ca053f 100644
> > --- a/arch/x86/include/asm/mce.h
> > +++ b/arch/x86/include/asm/mce.h
> > @@ -370,13 +370,24 @@ enum smca_bank_types {
> >  	SMCA_NBIO,	/* Northbridge IO Unit */
> >  	SMCA_PCIE,	/* PCI Express Unit */
> >  	SMCA_PCIE_V2,
> > +	SMCA_MPRAS,	/* MP for RAS */
> >  	SMCA_XGMI_PCS,	/* xGMI PCS Unit */
> > +	SMCA_SSBDCI,	/* Die to Die Interconnect */
> >  	SMCA_NBIF,	/* NBIF Unit */
> >  	SMCA_SHUB,	/* System HUB Unit */
> >  	SMCA_SATA,	/* SATA Unit */
> >  	SMCA_USB,	/* USB Unit */
> > +	SMCA_MPDACC,	/* MP for Data Acceleration */
> > +	SMCA_MPM,	/* Microprocessor Manageability Core */
> > +	SMCA_MPASP,	/* AMD Secure Processor */
> > +	SMCA_MPASP_V2,
> > +	SMCA_MPART,	/* AMD Root of Trust Microprocessor */
> > +	SMCA_DACC_FE,	/* Data Acceleration Front-end */
> > +	SMCA_DACC_BE,	/* Data Acceleration Back-end */
> >  	SMCA_USR_DP,	/* Ultra Short Reach Data Plane Controller */
> >  	SMCA_USR_CP,	/* Ultra Short Reach Control Plane Controller */
> > +	SMCA_EDDR5CMN,	/* eDDR5 CMN */
> > +	SMCA_PCIE_PL,	/* PCIe Link */
> >  	SMCA_GMI_PCS,	/* GMI PCS Unit */
> >  	SMCA_XGMI_PHY,	/* xGMI PHY Unit */
> >  	SMCA_WAFL_PHY,	/* WAFL PHY Unit */
> 
> That ordering looks rather random. Does it matter which enum number a bank
> type is? If not, let's sort them alphabetically for easier reading...
> 
> In the remaining places too.
> 

The ordering is based on the HWID_MCATYPE() tuple. The intent is to keep
those in numerical order for easy reference with documentation.

See: smca_hwid_mcatypes[]

Thanks,
Yazen