[PATCH v2 0/2] AMD SMCA Updates

[PATCH v2 0/2] AMD SMCA Updates

[Yazen Ghannam]

Hi all,

This set adds supports for SMCA changes in future AMD systems.

Patch 1 adds new bank types and error descriptions used in future AMD
systems.

Patch 2 adjusts how SMCA bank information is cached. Future AMD systems
will have different bank type layouts between logical CPUs. So having a
single system-wide cache of the layout won't be correct.

Patch 1 from V1 was dropped from this set. I plan to send another set
for the thresholding sysfs updates discussed here:
https://lkml.kernel.org/r/20211203020017.728440-2-yazen.ghannam@amd.com

I expect there will be a few patches for adding the new interface,
removing the old interface and stale data strucutures, and documenting
the new interface.

Thanks,
Yazen

Link:
https://lkml.kernel.org/r/20211203020017.728440-1-yazen.ghannam@amd.com

v1->v2:
* Drop patch 1 from v1 set and rework the remaining patches.

Yazen Ghannam (2):
  x86/MCE/AMD, EDAC/mce_amd: Add new SMCA Bank Types
  x86/MCE/AMD, EDAC/mce_amd: Support non-uniform MCA bank type
    enumeration

 arch/x86/include/asm/mce.h              |  25 ++--
 arch/x86/kernel/cpu/mce/amd.c           |  87 ++++++++++----
 drivers/edac/mce_amd.c                  | 146 +++++++++++++++++++++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c |   2 +-
 4 files changed, 202 insertions(+), 58 deletions(-)

-- 
2.25.1

[PATCH v2 1/2] x86/MCE/AMD, EDAC/mce_amd: Add new SMCA Bank Types

[Yazen Ghannam]

Add HWID and McaType values for new SMCA bank types, and add their error
descriptions to edac_mce_amd.

The "PHY" bank types all have the same error descriptions, and the NBIF
and SHUB bank types have the same error descriptions. So reuse the same
arrays where appropriate.

Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
---
Link:
https://lkml.kernel.org/r/20211203020017.728440-3-yazen.ghannam@amd.com

v1->v2:
* Rework due to dropping patch 1 from v1.

 arch/x86/include/asm/mce.h    |   7 ++
 arch/x86/kernel/cpu/mce/amd.c |  28 +++++++
 drivers/edac/mce_amd.c        | 135 ++++++++++++++++++++++++++++++++--
 3 files changed, 163 insertions(+), 7 deletions(-)

diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h
index d58f4f2e006f..52d2b35cac2d 100644
--- a/arch/x86/include/asm/mce.h
+++ b/arch/x86/include/asm/mce.h
@@ -313,12 +313,19 @@ enum smca_bank_types {
 	SMCA_SMU,	/* System Management Unit */
 	SMCA_SMU_V2,
 	SMCA_MP5,	/* Microprocessor 5 Unit */
+	SMCA_MPDMA,	/* MPDMA Unit */
 	SMCA_NBIO,	/* Northbridge IO Unit */
 	SMCA_PCIE,	/* PCI Express Unit */
 	SMCA_PCIE_V2,
 	SMCA_XGMI_PCS,	/* xGMI PCS Unit */
+	SMCA_NBIF,	/* NBIF Unit */
+	SMCA_SHUB,	/* System HUB Unit */
+	SMCA_SATA,	/* SATA Unit */
+	SMCA_USB,	/* USB Unit */
+	SMCA_GMI_PCS,	/* GMI PCS Unit */
 	SMCA_XGMI_PHY,	/* xGMI PHY Unit */
 	SMCA_WAFL_PHY,	/* WAFL PHY Unit */
+	SMCA_GMI_PHY,	/* GMI PHY Unit */
 	N_SMCA_BANK_TYPES
 };
 
diff --git a/arch/x86/kernel/cpu/mce/amd.c b/arch/x86/kernel/cpu/mce/amd.c
index 2f351838d5f7..ffe8ea8bc877 100644
--- a/arch/x86/kernel/cpu/mce/amd.c
+++ b/arch/x86/kernel/cpu/mce/amd.c
@@ -95,11 +95,18 @@ static struct smca_bank_name smca_names[] = {
 	[SMCA_PSP ... SMCA_PSP_V2]	= { "psp",		"Platform Security Processor" },
 	[SMCA_SMU ... SMCA_SMU_V2]	= { "smu",		"System Management Unit" },
 	[SMCA_MP5]			= { "mp5",		"Microprocessor 5 Unit" },
+	[SMCA_MPDMA]			= { "mpdma",		"MPDMA Unit" },
 	[SMCA_NBIO]			= { "nbio",		"Northbridge IO Unit" },
 	[SMCA_PCIE ... SMCA_PCIE_V2]	= { "pcie",		"PCI Express Unit" },
 	[SMCA_XGMI_PCS]			= { "xgmi_pcs",		"Ext Global Memory Interconnect PCS Unit" },
+	[SMCA_NBIF]			= { "nbif",		"NBIF Unit" },
+	[SMCA_SHUB]			= { "shub",		"System Hub Unit" },
+	[SMCA_SATA]			= { "sata",		"SATA Unit" },
+	[SMCA_USB]			= { "usb",		"USB Unit" },
+	[SMCA_GMI_PCS]			= { "gmi_pcs",		"Global Memory Interconnect PCS Unit" },
 	[SMCA_XGMI_PHY]			= { "xgmi_phy",		"Ext Global Memory Interconnect PHY Unit" },
 	[SMCA_WAFL_PHY]			= { "wafl_phy",		"WAFL PHY Unit" },
+	[SMCA_GMI_PHY]			= { "gmi_phy",		"Global Memory Interconnect PHY Unit" },
 };
 
 static const char *smca_get_name(enum smca_bank_types t)
@@ -174,6 +181,9 @@ static struct smca_hwid smca_hwid_mcatypes[] = {
 	/* Microprocessor 5 Unit MCA type */
 	{ SMCA_MP5,	 HWID_MCATYPE(0x01, 0x2)	},
 
+	/* MPDMA MCA type */
+	{ SMCA_MPDMA,	 HWID_MCATYPE(0x01, 0x3)	},
+
 	/* Northbridge IO Unit MCA type */
 	{ SMCA_NBIO,	 HWID_MCATYPE(0x18, 0x0)	},
 
@@ -184,11 +194,29 @@ static struct smca_hwid smca_hwid_mcatypes[] = {
 	/* xGMI PCS MCA type */
 	{ SMCA_XGMI_PCS, HWID_MCATYPE(0x50, 0x0)	},
 
+	/* NBIF MCA type */
+	{ SMCA_NBIF,	 HWID_MCATYPE(0x6C, 0x0)	},
+
+	/* SHUB MCA type */
+	{ SMCA_SHUB,	 HWID_MCATYPE(0x80, 0x0)	},
+
+	/* SATA MCA type */
+	{ SMCA_SATA,	 HWID_MCATYPE(0xA8, 0x0)	},
+
+	/* USB MCA type */
+	{ SMCA_USB,	 HWID_MCATYPE(0xAA, 0x0)	},
+
+	/* GMI PCS MCA type */
+	{ SMCA_GMI_PCS,  HWID_MCATYPE(0x241, 0x0)	},
+
 	/* xGMI PHY MCA type */
 	{ SMCA_XGMI_PHY, HWID_MCATYPE(0x259, 0x0)	},
 
 	/* WAFL PHY MCA type */
 	{ SMCA_WAFL_PHY, HWID_MCATYPE(0x267, 0x0)	},
+
+	/* GMI PHY MCA type */
+	{ SMCA_GMI_PHY,	 HWID_MCATYPE(0x269, 0x0)	},
 };
 
 struct smca_bank smca_banks[MAX_NR_BANKS];
diff --git a/drivers/edac/mce_amd.c b/drivers/edac/mce_amd.c
index 67dbf4c31271..cfd3f7ae9251 100644
--- a/drivers/edac/mce_amd.c
+++ b/drivers/edac/mce_amd.c
@@ -399,6 +399,63 @@ static const char * const smca_mp5_mce_desc[] = {
 	"Instruction Tag Cache Bank B ECC or parity error",
 };
 
+static const char * const smca_mpdma_mce_desc[] = {
+	"Main SRAM [31:0] bank ECC or parity error",
+	"Main SRAM [63:32] bank ECC or parity error",
+	"Main SRAM [95:64] bank ECC or parity error",
+	"Main SRAM [127:96] bank ECC or parity error",
+	"Data Cache Bank A ECC or parity error",
+	"Data Cache Bank B ECC or parity error",
+	"Data Tag Cache Bank A ECC or parity error",
+	"Data Tag Cache Bank B ECC or parity error",
+	"Instruction Cache Bank A ECC or parity error",
+	"Instruction Cache Bank B ECC or parity error",
+	"Instruction Tag Cache Bank A ECC or parity error",
+	"Instruction Tag Cache Bank B ECC or parity error",
+	"Data Cache Bank A ECC or parity error",
+	"Data Cache Bank B ECC or parity error",
+	"Data Tag Cache Bank A ECC or parity error",
+	"Data Tag Cache Bank B ECC or parity error",
+	"Instruction Cache Bank A ECC or parity error",
+	"Instruction Cache Bank B ECC or parity error",
+	"Instruction Tag Cache Bank A ECC or parity error",
+	"Instruction Tag Cache Bank B ECC or parity error",
+	"Data Cache Bank A ECC or parity error",
+	"Data Cache Bank B ECC or parity error",
+	"Data Tag Cache Bank A ECC or parity error",
+	"Data Tag Cache Bank B ECC or parity error",
+	"Instruction Cache Bank A ECC or parity error",
+	"Instruction Cache Bank B ECC or parity error",
+	"Instruction Tag Cache Bank A ECC or parity error",
+	"Instruction Tag Cache Bank B ECC or parity error",
+	"System Hub Read Buffer ECC or parity error",
+	"MPDMA TVF DVSEC Memory ECC or parity error",
+	"MPDMA TVF MMIO Mailbox0 ECC or parity error",
+	"MPDMA TVF MMIO Mailbox1 ECC or parity error",
+	"MPDMA TVF Doorbell Memory ECC or parity error",
+	"MPDMA TVF SDP Slave Memory 0 ECC or parity error",
+	"MPDMA TVF SDP Slave Memory 1 ECC or parity error",
+	"MPDMA TVF SDP Slave Memory 2 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 0 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 1 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 2 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 3 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 4 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 5 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 6 ECC or parity error",
+	"MPDMA PTE Command FIFO ECC or parity error",
+	"MPDMA PTE Hub Data FIFO ECC or parity error",
+	"MPDMA PTE Internal Data FIFO ECC or parity error",
+	"MPDMA PTE Command Memory DMA ECC or parity error",
+	"MPDMA PTE Command Memory Internal ECC or parity error",
+	"MPDMA PTE DMA Completion FIFO ECC or parity error",
+	"MPDMA PTE Tablewalk Completion FIFO ECC or parity error",
+	"MPDMA PTE Descriptor Completion FIFO ECC or parity error",
+	"MPDMA PTE ReadOnly Completion FIFO ECC or parity error",
+	"MPDMA PTE DirectWrite Completion FIFO ECC or parity error",
+	"SDP Watchdog Timer expired",
+};
+
 static const char * const smca_nbio_mce_desc[] = {
 	"ECC or Parity error",
 	"PCIE error",
@@ -448,7 +505,7 @@ static const char * const smca_xgmipcs_mce_desc[] = {
 	"Rx Replay Timeout Error",
 	"LinkSub Tx Timeout Error",
 	"LinkSub Rx Timeout Error",
-	"Rx CMD Pocket Error",
+	"Rx CMD Packet Error",
 };
 
 static const char * const smca_xgmiphy_mce_desc[] = {
@@ -458,11 +515,66 @@ static const char * const smca_xgmiphy_mce_desc[] = {
 	"PHY APB error",
 };
 
-static const char * const smca_waflphy_mce_desc[] = {
-	"RAM ECC Error",
-	"ARC instruction buffer parity error",
-	"ARC data buffer parity error",
-	"PHY APB error",
+static const char * const smca_nbif_mce_desc[] = {
+	"Timeout error from GMI",
+	"SRAM ECC error",
+	"NTB Error Event",
+	"SDP Parity error",
+};
+
+static const char * const smca_sata_mce_desc[] = {
+	"Parity error for port 0",
+	"Parity error for port 1",
+	"Parity error for port 2",
+	"Parity error for port 3",
+	"Parity error for port 4",
+	"Parity error for port 5",
+	"Parity error for port 6",
+	"Parity error for port 7",
+};
+
+static const char * const smca_usb_mce_desc[] = {
+	"Parity error or ECC error for S0 RAM0",
+	"Parity error or ECC error for S0 RAM1",
+	"Parity error or ECC error for S0 RAM2",
+	"Parity error for PHY RAM0",
+	"Parity error for PHY RAM1",
+	"AXI Slave Response error",
+};
+
+static const char * const smca_gmipcs_mce_desc[] = {
+	"Data Loss Error",
+	"Training Error",
+	"Replay Parity Error",
+	"Rx Fifo Underflow Error",
+	"Rx Fifo Overflow Error",
+	"CRC Error",
+	"BER Exceeded Error",
+	"Tx Fifo Underflow Error",
+	"Replay Buffer Parity Error",
+	"Tx Overflow Error",
+	"Replay Fifo Overflow Error",
+	"Replay Fifo Underflow Error",
+	"Elastic Fifo Overflow Error",
+	"Deskew Error",
+	"Offline Error",
+	"Data Startup Limit Error",
+	"FC Init Timeout Error",
+	"Recovery Timeout Error",
+	"Ready Serial Timeout Error",
+	"Ready Serial Attempt Error",
+	"Recovery Attempt Error",
+	"Recovery Relock Attempt Error",
+	"Deskew Abort Error",
+	"Rx Buffer Error",
+	"Rx LFDS Fifo Overflow Error",
+	"Rx LFDS Fifo Underflow Error",
+	"LinkSub Tx Timeout Error",
+	"LinkSub Rx Timeout Error",
+	"Rx CMD Packet Error",
+	"LFDS Training Timeout Error",
+	"LFDS FC Init Timeout Error",
+	"Data Loss Error",
 };
 
 struct smca_mce_desc {
@@ -490,12 +602,21 @@ static struct smca_mce_desc smca_mce_descs[] = {
 	[SMCA_SMU]	= { smca_smu_mce_desc,	ARRAY_SIZE(smca_smu_mce_desc)	},
 	[SMCA_SMU_V2]	= { smca_smu2_mce_desc,	ARRAY_SIZE(smca_smu2_mce_desc)	},
 	[SMCA_MP5]	= { smca_mp5_mce_desc,	ARRAY_SIZE(smca_mp5_mce_desc)	},
+	[SMCA_MPDMA]	= { smca_mpdma_mce_desc,	ARRAY_SIZE(smca_mpdma_mce_desc)	},
 	[SMCA_NBIO]	= { smca_nbio_mce_desc,	ARRAY_SIZE(smca_nbio_mce_desc)	},
 	[SMCA_PCIE]	= { smca_pcie_mce_desc,	ARRAY_SIZE(smca_pcie_mce_desc)	},
 	[SMCA_PCIE_V2]	= { smca_pcie2_mce_desc,   ARRAY_SIZE(smca_pcie2_mce_desc)	},
 	[SMCA_XGMI_PCS]	= { smca_xgmipcs_mce_desc, ARRAY_SIZE(smca_xgmipcs_mce_desc)	},
+	/* NBIF and SHUB have the same error descriptions, for now. */
+	[SMCA_NBIF]	= { smca_nbif_mce_desc, ARRAY_SIZE(smca_nbif_mce_desc)	},
+	[SMCA_SHUB]	= { smca_nbif_mce_desc, ARRAY_SIZE(smca_nbif_mce_desc)	},
+	[SMCA_SATA]	= { smca_sata_mce_desc, ARRAY_SIZE(smca_sata_mce_desc)	},
+	[SMCA_USB]	= { smca_usb_mce_desc,	ARRAY_SIZE(smca_usb_mce_desc)	},
+	[SMCA_GMI_PCS]	= { smca_gmipcs_mce_desc,  ARRAY_SIZE(smca_gmipcs_mce_desc)	},
+	/* All the PHY bank types have the same error descriptions, for now. */
 	[SMCA_XGMI_PHY]	= { smca_xgmiphy_mce_desc, ARRAY_SIZE(smca_xgmiphy_mce_desc)	},
-	[SMCA_WAFL_PHY]	= { smca_waflphy_mce_desc, ARRAY_SIZE(smca_waflphy_mce_desc)	},
+	[SMCA_WAFL_PHY]	= { smca_xgmiphy_mce_desc, ARRAY_SIZE(smca_xgmiphy_mce_desc)	},
+	[SMCA_GMI_PHY]	= { smca_xgmiphy_mce_desc, ARRAY_SIZE(smca_xgmiphy_mce_desc)	},
 };
 
 static bool f12h_mc0_mce(u16 ec, u8 xec)
-- 
2.25.1

[PATCH v2 2/2] x86/MCE/AMD, EDAC/mce_amd: Support non-uniform MCA bank type enumeration

[Yazen Ghannam]

AMD systems currently lay out MCA bank types such that the type of bank
number "i" is either the same across all CPUs or is Reserved/Read-as-Zero.

For example:

  Bank # | CPUx | CPUy
    0      LS     LS
    1      RAZ    UMC
    2      CS     CS
    3      SMU    RAZ

Future AMD systems will lay out MCA bank types such that the type of
bank number "i" may be different across CPUs.

For example:

  Bank # | CPUx | CPUy
    0      LS     LS
    1      RAZ    UMC
    2      CS     NBIO
    3      SMU    RAZ

Change the structures that cache MCA bank types to be per-CPU and update
smca_get_bank_type() to handle this change.

Move some SMCA-specific structures to amd.c from mce.h, since they no
longer need to be global.

Break out the "count" for bank types from struct smca_hwid, since this
should provide a per-CPU count rather than a system-wide count.

Apply the "const" qualifier to the struct smca_hwid_mcatypes array. The
values in this array should not change at runtime.

Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
---
Link:
https://lkml.kernel.org/r/20211203020017.728440-4-yazen.ghannam@amd.com

v1->v2:
* Rework due to dropping patch 1 from v1.

 arch/x86/include/asm/mce.h              | 18 +-------
 arch/x86/kernel/cpu/mce/amd.c           | 59 +++++++++++++++----------
 drivers/edac/mce_amd.c                  | 11 +----
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c |  2 +-
 4 files changed, 39 insertions(+), 51 deletions(-)

diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h
index 52d2b35cac2d..cc73061e7255 100644
--- a/arch/x86/include/asm/mce.h
+++ b/arch/x86/include/asm/mce.h
@@ -329,22 +329,6 @@ enum smca_bank_types {
 	N_SMCA_BANK_TYPES
 };
 
-#define HWID_MCATYPE(hwid, mcatype) (((hwid) << 16) | (mcatype))
-
-struct smca_hwid {
-	unsigned int bank_type;	/* Use with smca_bank_types for easy indexing. */
-	u32 hwid_mcatype;	/* (hwid,mcatype) tuple */
-	u8 count;		/* Number of instances. */
-};
-
-struct smca_bank {
-	struct smca_hwid *hwid;
-	u32 id;			/* Value of MCA_IPID[InstanceId]. */
-	u8 sysfs_id;		/* Value used for sysfs name. */
-};
-
-extern struct smca_bank smca_banks[MAX_NR_BANKS];
-
 extern const char *smca_get_long_name(enum smca_bank_types t);
 extern bool amd_mce_is_memory_error(struct mce *m);
 
@@ -352,7 +336,7 @@ extern int mce_threshold_create_device(unsigned int cpu);
 extern int mce_threshold_remove_device(unsigned int cpu);
 
 void mce_amd_feature_init(struct cpuinfo_x86 *c);
-enum smca_bank_types smca_get_bank_type(unsigned int bank);
+enum smca_bank_types smca_get_bank_type(unsigned int cpu, unsigned int bank);
 #else
 
 static inline int mce_threshold_create_device(unsigned int cpu)		{ return 0; };
diff --git a/arch/x86/kernel/cpu/mce/amd.c b/arch/x86/kernel/cpu/mce/amd.c
index ffe8ea8bc877..345ccdf329bc 100644
--- a/arch/x86/kernel/cpu/mce/amd.c
+++ b/arch/x86/kernel/cpu/mce/amd.c
@@ -71,6 +71,22 @@ static const char * const smca_umc_block_names[] = {
 	"misc_umc"
 };
 
+#define HWID_MCATYPE(hwid, mcatype) (((hwid) << 16) | (mcatype))
+
+struct smca_hwid {
+	unsigned int bank_type;	/* Use with smca_bank_types for easy indexing. */
+	u32 hwid_mcatype;	/* (hwid,mcatype) tuple */
+};
+
+struct smca_bank {
+	const struct smca_hwid *hwid;
+	u32 id;			/* Value of MCA_IPID[InstanceId]. */
+	u8 sysfs_id;		/* Value used for sysfs name. */
+};
+
+static DEFINE_PER_CPU_READ_MOSTLY(struct smca_bank[MAX_NR_BANKS], smca_banks);
+static DEFINE_PER_CPU_READ_MOSTLY(u8[N_SMCA_BANK_TYPES], smca_bank_counts);
+
 struct smca_bank_name {
 	const char *name;	/* Short name for sysfs */
 	const char *long_name;	/* Long name for pretty-printing */
@@ -126,14 +142,14 @@ const char *smca_get_long_name(enum smca_bank_types t)
 }
 EXPORT_SYMBOL_GPL(smca_get_long_name);
 
-enum smca_bank_types smca_get_bank_type(unsigned int bank)
+enum smca_bank_types smca_get_bank_type(unsigned int cpu, unsigned int bank)
 {
 	struct smca_bank *b;
 
 	if (bank >= MAX_NR_BANKS)
 		return N_SMCA_BANK_TYPES;
 
-	b = &smca_banks[bank];
+	b = &per_cpu(smca_banks, cpu)[bank];
 	if (!b->hwid)
 		return N_SMCA_BANK_TYPES;
 
@@ -141,7 +157,7 @@ enum smca_bank_types smca_get_bank_type(unsigned int bank)
 }
 EXPORT_SYMBOL_GPL(smca_get_bank_type);
 
-static struct smca_hwid smca_hwid_mcatypes[] = {
+static const struct smca_hwid smca_hwid_mcatypes[] = {
 	/* { bank_type, hwid_mcatype } */
 
 	/* Reserved type */
@@ -219,9 +235,6 @@ static struct smca_hwid smca_hwid_mcatypes[] = {
 	{ SMCA_GMI_PHY,	 HWID_MCATYPE(0x269, 0x0)	},
 };
 
-struct smca_bank smca_banks[MAX_NR_BANKS];
-EXPORT_SYMBOL_GPL(smca_banks);
-
 /*
  * In SMCA enabled processors, we can have multiple banks for a given IP type.
  * So to define a unique name for each bank, we use a temp c-string to append
@@ -277,8 +290,9 @@ static void smca_set_misc_banks_map(unsigned int bank, unsigned int cpu)
 
 static void smca_configure(unsigned int bank, unsigned int cpu)
 {
+	u8 *bank_counts = this_cpu_ptr(smca_bank_counts);
+	const struct smca_hwid *s_hwid;
 	unsigned int i, hwid_mcatype;
-	struct smca_hwid *s_hwid;
 	u32 high, low;
 	u32 smca_config = MSR_AMD64_SMCA_MCx_CONFIG(bank);
 
@@ -314,10 +328,6 @@ static void smca_configure(unsigned int bank, unsigned int cpu)
 
 	smca_set_misc_banks_map(bank, cpu);
 
-	/* Return early if this bank was already initialized. */
-	if (smca_banks[bank].hwid && smca_banks[bank].hwid->hwid_mcatype != 0)
-		return;
-
 	if (rdmsr_safe(MSR_AMD64_SMCA_MCx_IPID(bank), &low, &high)) {
 		pr_warn("Failed to read MCA_IPID for bank %d\n", bank);
 		return;
@@ -328,10 +338,11 @@ static void smca_configure(unsigned int bank, unsigned int cpu)
 
 	for (i = 0; i < ARRAY_SIZE(smca_hwid_mcatypes); i++) {
 		s_hwid = &smca_hwid_mcatypes[i];
+
 		if (hwid_mcatype == s_hwid->hwid_mcatype) {
-			smca_banks[bank].hwid = s_hwid;
-			smca_banks[bank].id = low;
-			smca_banks[bank].sysfs_id = s_hwid->count++;
+			this_cpu_ptr(smca_banks)[bank].hwid = s_hwid;
+			this_cpu_ptr(smca_banks)[bank].id = low;
+			this_cpu_ptr(smca_banks)[bank].sysfs_id = bank_counts[s_hwid->bank_type]++;
 			break;
 		}
 	}
@@ -617,7 +628,7 @@ prepare_threshold_block(unsigned int bank, unsigned int block, u32 addr,
 
 bool amd_filter_mce(struct mce *m)
 {
-	enum smca_bank_types bank_type = smca_get_bank_type(m->bank);
+	enum smca_bank_types bank_type = smca_get_bank_type(m->extcpu, m->bank);
 	struct cpuinfo_x86 *c = &boot_cpu_data;
 
 	/* See Family 17h Models 10h-2Fh Erratum #1114. */
@@ -655,7 +666,7 @@ static void disable_err_thresholding(struct cpuinfo_x86 *c, unsigned int bank)
 	} else if (c->x86 == 0x17 &&
 		   (c->x86_model >= 0x10 && c->x86_model <= 0x2F)) {
 
-		if (smca_get_bank_type(bank) != SMCA_IF)
+		if (smca_get_bank_type(smp_processor_id(), bank) != SMCA_IF)
 			return;
 
 		msrs[0] = MSR_AMD64_SMCA_MCx_MISC(bank);
@@ -723,7 +734,7 @@ bool amd_mce_is_memory_error(struct mce *m)
 	u8 xec = (m->status >> 16) & 0x1f;
 
 	if (mce_flags.smca)
-		return smca_get_bank_type(m->bank) == SMCA_UMC && xec == 0x0;
+		return smca_get_bank_type(m->extcpu, m->bank) == SMCA_UMC && xec == 0x0;
 
 	return m->bank == 4 && xec == 0x8;
 }
@@ -1039,7 +1050,7 @@ static struct kobj_type threshold_ktype = {
 	.release		= threshold_block_release,
 };
 
-static const char *get_name(unsigned int bank, struct threshold_block *b)
+static const char *get_name(unsigned int cpu, unsigned int bank, struct threshold_block *b)
 {
 	enum smca_bank_types bank_type;
 
@@ -1050,7 +1061,7 @@ static const char *get_name(unsigned int bank, struct threshold_block *b)
 		return th_names[bank];
 	}
 
-	bank_type = smca_get_bank_type(bank);
+	bank_type = smca_get_bank_type(cpu, bank);
 	if (bank_type >= N_SMCA_BANK_TYPES)
 		return NULL;
 
@@ -1060,12 +1071,12 @@ static const char *get_name(unsigned int bank, struct threshold_block *b)
 		return NULL;
 	}
 
-	if (smca_banks[bank].hwid->count == 1)
+	if (per_cpu(smca_bank_counts, cpu)[bank_type] == 1)
 		return smca_get_name(bank_type);
 
 	snprintf(buf_mcatype, MAX_MCATYPE_NAME_LEN,
-		 "%s_%x", smca_get_name(bank_type),
-			  smca_banks[bank].sysfs_id);
+		 "%s_%u", smca_get_name(bank_type),
+			  per_cpu(smca_banks, cpu)[bank].sysfs_id);
 	return buf_mcatype;
 }
 
@@ -1121,7 +1132,7 @@ static int allocate_threshold_blocks(unsigned int cpu, struct threshold_bank *tb
 	else
 		tb->blocks = b;
 
-	err = kobject_init_and_add(&b->kobj, &threshold_ktype, tb->kobj, get_name(bank, b));
+	err = kobject_init_and_add(&b->kobj, &threshold_ktype, tb->kobj, get_name(cpu, bank, b));
 	if (err)
 		goto out_free;
 recurse:
@@ -1176,7 +1187,7 @@ static int threshold_create_bank(struct threshold_bank **bp, unsigned int cpu,
 	struct device *dev = this_cpu_read(mce_device);
 	struct amd_northbridge *nb = NULL;
 	struct threshold_bank *b = NULL;
-	const char *name = get_name(bank, NULL);
+	const char *name = get_name(cpu, bank, NULL);
 	int err = 0;
 
 	if (!dev)
diff --git a/drivers/edac/mce_amd.c b/drivers/edac/mce_amd.c
index cfd3f7ae9251..cc5c63feb26a 100644
--- a/drivers/edac/mce_amd.c
+++ b/drivers/edac/mce_amd.c
@@ -1166,20 +1166,13 @@ static void decode_mc6_mce(struct mce *m)
 /* Decode errors according to Scalable MCA specification */
 static void decode_smca_error(struct mce *m)
 {
-	struct smca_hwid *hwid;
-	enum smca_bank_types bank_type;
+	enum smca_bank_types bank_type = smca_get_bank_type(m->extcpu, m->bank);
 	const char *ip_name;
 	u8 xec = XEC(m->status, xec_mask);
 
-	if (m->bank >= ARRAY_SIZE(smca_banks))
+	if (bank_type >= N_SMCA_BANK_TYPES)
 		return;
 
-	hwid = smca_banks[m->bank].hwid;
-	if (!hwid)
-		return;
-
-	bank_type = hwid->bank_type;
-
 	if (bank_type == SMCA_RESERVED) {
 		pr_emerg(HW_ERR "Bank %d is reserved.\n", m->bank);
 		return;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 08133de21fdd..75dad0214dc7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -2647,7 +2647,7 @@ static int amdgpu_bad_page_notifier(struct notifier_block *nb,
 	 * and error occurred in DramECC (Extended error code = 0) then only
 	 * process the error, else bail out.
 	 */
-	if (!m || !((smca_get_bank_type(m->bank) == SMCA_UMC_V2) &&
+	if (!m || !((smca_get_bank_type(m->extcpu, m->bank) == SMCA_UMC_V2) &&
 		    (XEC(m->status, 0x3f) == 0x0)))
 		return NOTIFY_DONE;
 
-- 
2.25.1

[tip: ras/core] x86/MCE/AMD, EDAC/mce_amd: Add new SMCA bank types

[tip-bot2 for Yazen Ghannam]

The following commit has been merged into the ras/core branch of tip:

Commit-ID:     5176a93ab27aef1b9f4496fc68e6c303a011d7cc
Gitweb:        https://git.kernel.org/tip/5176a93ab27aef1b9f4496fc68e6c303a011d7cc
Author:        Yazen Ghannam <yazen.ghannam@amd.com>
AuthorDate:    Thu, 16 Dec 2021 16:29:04 
Committer:     Borislav Petkov <bp@suse.de>
CommitterDate: Wed, 22 Dec 2021 17:19:18 +01:00

x86/MCE/AMD, EDAC/mce_amd: Add new SMCA bank types

Add HWID and McaType values for new SMCA bank types, and add their error
descriptions to edac_mce_amd.

The "PHY" bank types all have the same error descriptions, and the NBIF
and SHUB bank types have the same error descriptions. So reuse the same
arrays where appropriate.

  [ bp: Remove useless comments over hwid types. ]

Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/20211216162905.4132657-2-yazen.ghannam@amd.com
---
 arch/x86/include/asm/mce.h    |   7 ++-
 arch/x86/kernel/cpu/mce/amd.c |  21 +++--
 drivers/edac/mce_amd.c        | 135 +++++++++++++++++++++++++++++++--
 3 files changed, 151 insertions(+), 12 deletions(-)

diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h
index d58f4f2..52d2b35 100644
--- a/arch/x86/include/asm/mce.h
+++ b/arch/x86/include/asm/mce.h
@@ -313,12 +313,19 @@ enum smca_bank_types {
 	SMCA_SMU,	/* System Management Unit */
 	SMCA_SMU_V2,
 	SMCA_MP5,	/* Microprocessor 5 Unit */
+	SMCA_MPDMA,	/* MPDMA Unit */
 	SMCA_NBIO,	/* Northbridge IO Unit */
 	SMCA_PCIE,	/* PCI Express Unit */
 	SMCA_PCIE_V2,
 	SMCA_XGMI_PCS,	/* xGMI PCS Unit */
+	SMCA_NBIF,	/* NBIF Unit */
+	SMCA_SHUB,	/* System HUB Unit */
+	SMCA_SATA,	/* SATA Unit */
+	SMCA_USB,	/* USB Unit */
+	SMCA_GMI_PCS,	/* GMI PCS Unit */
 	SMCA_XGMI_PHY,	/* xGMI PHY Unit */
 	SMCA_WAFL_PHY,	/* WAFL PHY Unit */
+	SMCA_GMI_PHY,	/* GMI PHY Unit */
 	N_SMCA_BANK_TYPES
 };
 
diff --git a/arch/x86/kernel/cpu/mce/amd.c b/arch/x86/kernel/cpu/mce/amd.c
index 2f35183..9407cdc 100644
--- a/arch/x86/kernel/cpu/mce/amd.c
+++ b/arch/x86/kernel/cpu/mce/amd.c
@@ -95,11 +95,18 @@ static struct smca_bank_name smca_names[] = {
 	[SMCA_PSP ... SMCA_PSP_V2]	= { "psp",		"Platform Security Processor" },
 	[SMCA_SMU ... SMCA_SMU_V2]	= { "smu",		"System Management Unit" },
 	[SMCA_MP5]			= { "mp5",		"Microprocessor 5 Unit" },
+	[SMCA_MPDMA]			= { "mpdma",		"MPDMA Unit" },
 	[SMCA_NBIO]			= { "nbio",		"Northbridge IO Unit" },
 	[SMCA_PCIE ... SMCA_PCIE_V2]	= { "pcie",		"PCI Express Unit" },
 	[SMCA_XGMI_PCS]			= { "xgmi_pcs",		"Ext Global Memory Interconnect PCS Unit" },
+	[SMCA_NBIF]			= { "nbif",		"NBIF Unit" },
+	[SMCA_SHUB]			= { "shub",		"System Hub Unit" },
+	[SMCA_SATA]			= { "sata",		"SATA Unit" },
+	[SMCA_USB]			= { "usb",		"USB Unit" },
+	[SMCA_GMI_PCS]			= { "gmi_pcs",		"Global Memory Interconnect PCS Unit" },
 	[SMCA_XGMI_PHY]			= { "xgmi_phy",		"Ext Global Memory Interconnect PHY Unit" },
 	[SMCA_WAFL_PHY]			= { "wafl_phy",		"WAFL PHY Unit" },
+	[SMCA_GMI_PHY]			= { "gmi_phy",		"Global Memory Interconnect PHY Unit" },
 };
 
 static const char *smca_get_name(enum smca_bank_types t)
@@ -174,6 +181,9 @@ static struct smca_hwid smca_hwid_mcatypes[] = {
 	/* Microprocessor 5 Unit MCA type */
 	{ SMCA_MP5,	 HWID_MCATYPE(0x01, 0x2)	},
 
+	/* MPDMA MCA type */
+	{ SMCA_MPDMA,	 HWID_MCATYPE(0x01, 0x3)	},
+
 	/* Northbridge IO Unit MCA type */
 	{ SMCA_NBIO,	 HWID_MCATYPE(0x18, 0x0)	},
 
@@ -181,14 +191,15 @@ static struct smca_hwid smca_hwid_mcatypes[] = {
 	{ SMCA_PCIE,	 HWID_MCATYPE(0x46, 0x0)	},
 	{ SMCA_PCIE_V2,	 HWID_MCATYPE(0x46, 0x1)	},
 
-	/* xGMI PCS MCA type */
 	{ SMCA_XGMI_PCS, HWID_MCATYPE(0x50, 0x0)	},
-
-	/* xGMI PHY MCA type */
+	{ SMCA_NBIF,	 HWID_MCATYPE(0x6C, 0x0)	},
+	{ SMCA_SHUB,	 HWID_MCATYPE(0x80, 0x0)	},
+	{ SMCA_SATA,	 HWID_MCATYPE(0xA8, 0x0)	},
+	{ SMCA_USB,	 HWID_MCATYPE(0xAA, 0x0)	},
+	{ SMCA_GMI_PCS,  HWID_MCATYPE(0x241, 0x0)	},
 	{ SMCA_XGMI_PHY, HWID_MCATYPE(0x259, 0x0)	},
-
-	/* WAFL PHY MCA type */
 	{ SMCA_WAFL_PHY, HWID_MCATYPE(0x267, 0x0)	},
+	{ SMCA_GMI_PHY,	 HWID_MCATYPE(0x269, 0x0)	},
 };
 
 struct smca_bank smca_banks[MAX_NR_BANKS];
diff --git a/drivers/edac/mce_amd.c b/drivers/edac/mce_amd.c
index 67dbf4c..cfd3f7a 100644
--- a/drivers/edac/mce_amd.c
+++ b/drivers/edac/mce_amd.c
@@ -399,6 +399,63 @@ static const char * const smca_mp5_mce_desc[] = {
 	"Instruction Tag Cache Bank B ECC or parity error",
 };
 
+static const char * const smca_mpdma_mce_desc[] = {
+	"Main SRAM [31:0] bank ECC or parity error",
+	"Main SRAM [63:32] bank ECC or parity error",
+	"Main SRAM [95:64] bank ECC or parity error",
+	"Main SRAM [127:96] bank ECC or parity error",
+	"Data Cache Bank A ECC or parity error",
+	"Data Cache Bank B ECC or parity error",
+	"Data Tag Cache Bank A ECC or parity error",
+	"Data Tag Cache Bank B ECC or parity error",
+	"Instruction Cache Bank A ECC or parity error",
+	"Instruction Cache Bank B ECC or parity error",
+	"Instruction Tag Cache Bank A ECC or parity error",
+	"Instruction Tag Cache Bank B ECC or parity error",
+	"Data Cache Bank A ECC or parity error",
+	"Data Cache Bank B ECC or parity error",
+	"Data Tag Cache Bank A ECC or parity error",
+	"Data Tag Cache Bank B ECC or parity error",
+	"Instruction Cache Bank A ECC or parity error",
+	"Instruction Cache Bank B ECC or parity error",
+	"Instruction Tag Cache Bank A ECC or parity error",
+	"Instruction Tag Cache Bank B ECC or parity error",
+	"Data Cache Bank A ECC or parity error",
+	"Data Cache Bank B ECC or parity error",
+	"Data Tag Cache Bank A ECC or parity error",
+	"Data Tag Cache Bank B ECC or parity error",
+	"Instruction Cache Bank A ECC or parity error",
+	"Instruction Cache Bank B ECC or parity error",
+	"Instruction Tag Cache Bank A ECC or parity error",
+	"Instruction Tag Cache Bank B ECC or parity error",
+	"System Hub Read Buffer ECC or parity error",
+	"MPDMA TVF DVSEC Memory ECC or parity error",
+	"MPDMA TVF MMIO Mailbox0 ECC or parity error",
+	"MPDMA TVF MMIO Mailbox1 ECC or parity error",
+	"MPDMA TVF Doorbell Memory ECC or parity error",
+	"MPDMA TVF SDP Slave Memory 0 ECC or parity error",
+	"MPDMA TVF SDP Slave Memory 1 ECC or parity error",
+	"MPDMA TVF SDP Slave Memory 2 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 0 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 1 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 2 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 3 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 4 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 5 ECC or parity error",
+	"MPDMA TVF SDP Master Memory 6 ECC or parity error",
+	"MPDMA PTE Command FIFO ECC or parity error",
+	"MPDMA PTE Hub Data FIFO ECC or parity error",
+	"MPDMA PTE Internal Data FIFO ECC or parity error",
+	"MPDMA PTE Command Memory DMA ECC or parity error",
+	"MPDMA PTE Command Memory Internal ECC or parity error",
+	"MPDMA PTE DMA Completion FIFO ECC or parity error",
+	"MPDMA PTE Tablewalk Completion FIFO ECC or parity error",
+	"MPDMA PTE Descriptor Completion FIFO ECC or parity error",
+	"MPDMA PTE ReadOnly Completion FIFO ECC or parity error",
+	"MPDMA PTE DirectWrite Completion FIFO ECC or parity error",
+	"SDP Watchdog Timer expired",
+};
+
 static const char * const smca_nbio_mce_desc[] = {
 	"ECC or Parity error",
 	"PCIE error",
@@ -448,7 +505,7 @@ static const char * const smca_xgmipcs_mce_desc[] = {
 	"Rx Replay Timeout Error",
 	"LinkSub Tx Timeout Error",
 	"LinkSub Rx Timeout Error",
-	"Rx CMD Pocket Error",
+	"Rx CMD Packet Error",
 };
 
 static const char * const smca_xgmiphy_mce_desc[] = {
@@ -458,11 +515,66 @@ static const char * const smca_xgmiphy_mce_desc[] = {
 	"PHY APB error",
 };
 
-static const char * const smca_waflphy_mce_desc[] = {
-	"RAM ECC Error",
-	"ARC instruction buffer parity error",
-	"ARC data buffer parity error",
-	"PHY APB error",
+static const char * const smca_nbif_mce_desc[] = {
+	"Timeout error from GMI",
+	"SRAM ECC error",
+	"NTB Error Event",
+	"SDP Parity error",
+};
+
+static const char * const smca_sata_mce_desc[] = {
+	"Parity error for port 0",
+	"Parity error for port 1",
+	"Parity error for port 2",
+	"Parity error for port 3",
+	"Parity error for port 4",
+	"Parity error for port 5",
+	"Parity error for port 6",
+	"Parity error for port 7",
+};
+
+static const char * const smca_usb_mce_desc[] = {
+	"Parity error or ECC error for S0 RAM0",
+	"Parity error or ECC error for S0 RAM1",
+	"Parity error or ECC error for S0 RAM2",
+	"Parity error for PHY RAM0",
+	"Parity error for PHY RAM1",
+	"AXI Slave Response error",
+};
+
+static const char * const smca_gmipcs_mce_desc[] = {
+	"Data Loss Error",
+	"Training Error",
+	"Replay Parity Error",
+	"Rx Fifo Underflow Error",
+	"Rx Fifo Overflow Error",
+	"CRC Error",
+	"BER Exceeded Error",
+	"Tx Fifo Underflow Error",
+	"Replay Buffer Parity Error",
+	"Tx Overflow Error",
+	"Replay Fifo Overflow Error",
+	"Replay Fifo Underflow Error",
+	"Elastic Fifo Overflow Error",
+	"Deskew Error",
+	"Offline Error",
+	"Data Startup Limit Error",
+	"FC Init Timeout Error",
+	"Recovery Timeout Error",
+	"Ready Serial Timeout Error",
+	"Ready Serial Attempt Error",
+	"Recovery Attempt Error",
+	"Recovery Relock Attempt Error",
+	"Deskew Abort Error",
+	"Rx Buffer Error",
+	"Rx LFDS Fifo Overflow Error",
+	"Rx LFDS Fifo Underflow Error",
+	"LinkSub Tx Timeout Error",
+	"LinkSub Rx Timeout Error",
+	"Rx CMD Packet Error",
+	"LFDS Training Timeout Error",
+	"LFDS FC Init Timeout Error",
+	"Data Loss Error",
 };
 
 struct smca_mce_desc {
@@ -490,12 +602,21 @@ static struct smca_mce_desc smca_mce_descs[] = {
 	[SMCA_SMU]	= { smca_smu_mce_desc,	ARRAY_SIZE(smca_smu_mce_desc)	},
 	[SMCA_SMU_V2]	= { smca_smu2_mce_desc,	ARRAY_SIZE(smca_smu2_mce_desc)	},
 	[SMCA_MP5]	= { smca_mp5_mce_desc,	ARRAY_SIZE(smca_mp5_mce_desc)	},
+	[SMCA_MPDMA]	= { smca_mpdma_mce_desc,	ARRAY_SIZE(smca_mpdma_mce_desc)	},
 	[SMCA_NBIO]	= { smca_nbio_mce_desc,	ARRAY_SIZE(smca_nbio_mce_desc)	},
 	[SMCA_PCIE]	= { smca_pcie_mce_desc,	ARRAY_SIZE(smca_pcie_mce_desc)	},
 	[SMCA_PCIE_V2]	= { smca_pcie2_mce_desc,   ARRAY_SIZE(smca_pcie2_mce_desc)	},
 	[SMCA_XGMI_PCS]	= { smca_xgmipcs_mce_desc, ARRAY_SIZE(smca_xgmipcs_mce_desc)	},
+	/* NBIF and SHUB have the same error descriptions, for now. */
+	[SMCA_NBIF]	= { smca_nbif_mce_desc, ARRAY_SIZE(smca_nbif_mce_desc)	},
+	[SMCA_SHUB]	= { smca_nbif_mce_desc, ARRAY_SIZE(smca_nbif_mce_desc)	},
+	[SMCA_SATA]	= { smca_sata_mce_desc, ARRAY_SIZE(smca_sata_mce_desc)	},
+	[SMCA_USB]	= { smca_usb_mce_desc,	ARRAY_SIZE(smca_usb_mce_desc)	},
+	[SMCA_GMI_PCS]	= { smca_gmipcs_mce_desc,  ARRAY_SIZE(smca_gmipcs_mce_desc)	},
+	/* All the PHY bank types have the same error descriptions, for now. */
 	[SMCA_XGMI_PHY]	= { smca_xgmiphy_mce_desc, ARRAY_SIZE(smca_xgmiphy_mce_desc)	},
-	[SMCA_WAFL_PHY]	= { smca_waflphy_mce_desc, ARRAY_SIZE(smca_waflphy_mce_desc)	},
+	[SMCA_WAFL_PHY]	= { smca_xgmiphy_mce_desc, ARRAY_SIZE(smca_xgmiphy_mce_desc)	},
+	[SMCA_GMI_PHY]	= { smca_xgmiphy_mce_desc, ARRAY_SIZE(smca_xgmiphy_mce_desc)	},
 };
 
 static bool f12h_mc0_mce(u16 ec, u8 xec)

[tip: ras/core] x86/MCE/AMD, EDAC/mce_amd: Support non-uniform MCA bank type enumeration

[tip-bot2 for Yazen Ghannam]

The following commit has been merged into the ras/core branch of tip:

Commit-ID:     91f75eb481cfaee5c4ed8fb5214bf2fbfa04bd7b
Gitweb:        https://git.kernel.org/tip/91f75eb481cfaee5c4ed8fb5214bf2fbfa04bd7b
Author:        Yazen Ghannam <yazen.ghannam@amd.com>
AuthorDate:    Thu, 16 Dec 2021 16:29:05 
Committer:     Borislav Petkov <bp@suse.de>
CommitterDate: Wed, 22 Dec 2021 17:22:09 +01:00

x86/MCE/AMD, EDAC/mce_amd: Support non-uniform MCA bank type enumeration

AMD systems currently lay out MCA bank types such that the type of bank
number "i" is either the same across all CPUs or is Reserved/Read-as-Zero.

For example:

  Bank # | CPUx | CPUy
    0      LS     LS
    1      RAZ    UMC
    2      CS     CS
    3      SMU    RAZ

Future AMD systems will lay out MCA bank types such that the type of
bank number "i" may be different across CPUs.

For example:

  Bank # | CPUx | CPUy
    0      LS     LS
    1      RAZ    UMC
    2      CS     NBIO
    3      SMU    RAZ

Change the structures that cache MCA bank types to be per-CPU and update
smca_get_bank_type() to handle this change.

Move some SMCA-specific structures to amd.c from mce.h, since they no
longer need to be global.

Break out the "count" for bank types from struct smca_hwid, since this
should provide a per-CPU count rather than a system-wide count.

Apply the "const" qualifier to the struct smca_hwid_mcatypes array. The
values in this array should not change at runtime.

Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/20211216162905.4132657-3-yazen.ghannam@amd.com
---
 arch/x86/include/asm/mce.h              | 18 +-------
 arch/x86/kernel/cpu/mce/amd.c           | 59 ++++++++++++++----------
 drivers/edac/mce_amd.c                  | 11 +----
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c |  2 +-
 4 files changed, 39 insertions(+), 51 deletions(-)

diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h
index 52d2b35..cc73061 100644
--- a/arch/x86/include/asm/mce.h
+++ b/arch/x86/include/asm/mce.h
@@ -329,22 +329,6 @@ enum smca_bank_types {
 	N_SMCA_BANK_TYPES
 };
 
-#define HWID_MCATYPE(hwid, mcatype) (((hwid) << 16) | (mcatype))
-
-struct smca_hwid {
-	unsigned int bank_type;	/* Use with smca_bank_types for easy indexing. */
-	u32 hwid_mcatype;	/* (hwid,mcatype) tuple */
-	u8 count;		/* Number of instances. */
-};
-
-struct smca_bank {
-	struct smca_hwid *hwid;
-	u32 id;			/* Value of MCA_IPID[InstanceId]. */
-	u8 sysfs_id;		/* Value used for sysfs name. */
-};
-
-extern struct smca_bank smca_banks[MAX_NR_BANKS];
-
 extern const char *smca_get_long_name(enum smca_bank_types t);
 extern bool amd_mce_is_memory_error(struct mce *m);
 
@@ -352,7 +336,7 @@ extern int mce_threshold_create_device(unsigned int cpu);
 extern int mce_threshold_remove_device(unsigned int cpu);
 
 void mce_amd_feature_init(struct cpuinfo_x86 *c);
-enum smca_bank_types smca_get_bank_type(unsigned int bank);
+enum smca_bank_types smca_get_bank_type(unsigned int cpu, unsigned int bank);
 #else
 
 static inline int mce_threshold_create_device(unsigned int cpu)		{ return 0; };
diff --git a/arch/x86/kernel/cpu/mce/amd.c b/arch/x86/kernel/cpu/mce/amd.c
index 9407cdc..a1e2f41 100644
--- a/arch/x86/kernel/cpu/mce/amd.c
+++ b/arch/x86/kernel/cpu/mce/amd.c
@@ -71,6 +71,22 @@ static const char * const smca_umc_block_names[] = {
 	"misc_umc"
 };
 
+#define HWID_MCATYPE(hwid, mcatype) (((hwid) << 16) | (mcatype))
+
+struct smca_hwid {
+	unsigned int bank_type;	/* Use with smca_bank_types for easy indexing. */
+	u32 hwid_mcatype;	/* (hwid,mcatype) tuple */
+};
+
+struct smca_bank {
+	const struct smca_hwid *hwid;
+	u32 id;			/* Value of MCA_IPID[InstanceId]. */
+	u8 sysfs_id;		/* Value used for sysfs name. */
+};
+
+static DEFINE_PER_CPU_READ_MOSTLY(struct smca_bank[MAX_NR_BANKS], smca_banks);
+static DEFINE_PER_CPU_READ_MOSTLY(u8[N_SMCA_BANK_TYPES], smca_bank_counts);
+
 struct smca_bank_name {
 	const char *name;	/* Short name for sysfs */
 	const char *long_name;	/* Long name for pretty-printing */
@@ -126,14 +142,14 @@ const char *smca_get_long_name(enum smca_bank_types t)
 }
 EXPORT_SYMBOL_GPL(smca_get_long_name);
 
-enum smca_bank_types smca_get_bank_type(unsigned int bank)
+enum smca_bank_types smca_get_bank_type(unsigned int cpu, unsigned int bank)
 {
 	struct smca_bank *b;
 
 	if (bank >= MAX_NR_BANKS)
 		return N_SMCA_BANK_TYPES;
 
-	b = &smca_banks[bank];
+	b = &per_cpu(smca_banks, cpu)[bank];
 	if (!b->hwid)
 		return N_SMCA_BANK_TYPES;
 
@@ -141,7 +157,7 @@ enum smca_bank_types smca_get_bank_type(unsigned int bank)
 }
 EXPORT_SYMBOL_GPL(smca_get_bank_type);
 
-static struct smca_hwid smca_hwid_mcatypes[] = {
+static const struct smca_hwid smca_hwid_mcatypes[] = {
 	/* { bank_type, hwid_mcatype } */
 
 	/* Reserved type */
@@ -202,9 +218,6 @@ static struct smca_hwid smca_hwid_mcatypes[] = {
 	{ SMCA_GMI_PHY,	 HWID_MCATYPE(0x269, 0x0)	},
 };
 
-struct smca_bank smca_banks[MAX_NR_BANKS];
-EXPORT_SYMBOL_GPL(smca_banks);
-
 /*
  * In SMCA enabled processors, we can have multiple banks for a given IP type.
  * So to define a unique name for each bank, we use a temp c-string to append
@@ -260,8 +273,9 @@ static void smca_set_misc_banks_map(unsigned int bank, unsigned int cpu)
 
 static void smca_configure(unsigned int bank, unsigned int cpu)
 {
+	u8 *bank_counts = this_cpu_ptr(smca_bank_counts);
+	const struct smca_hwid *s_hwid;
 	unsigned int i, hwid_mcatype;
-	struct smca_hwid *s_hwid;
 	u32 high, low;
 	u32 smca_config = MSR_AMD64_SMCA_MCx_CONFIG(bank);
 
@@ -297,10 +311,6 @@ static void smca_configure(unsigned int bank, unsigned int cpu)
 
 	smca_set_misc_banks_map(bank, cpu);
 
-	/* Return early if this bank was already initialized. */
-	if (smca_banks[bank].hwid && smca_banks[bank].hwid->hwid_mcatype != 0)
-		return;
-
 	if (rdmsr_safe(MSR_AMD64_SMCA_MCx_IPID(bank), &low, &high)) {
 		pr_warn("Failed to read MCA_IPID for bank %d\n", bank);
 		return;
@@ -311,10 +321,11 @@ static void smca_configure(unsigned int bank, unsigned int cpu)
 
 	for (i = 0; i < ARRAY_SIZE(smca_hwid_mcatypes); i++) {
 		s_hwid = &smca_hwid_mcatypes[i];
+
 		if (hwid_mcatype == s_hwid->hwid_mcatype) {
-			smca_banks[bank].hwid = s_hwid;
-			smca_banks[bank].id = low;
-			smca_banks[bank].sysfs_id = s_hwid->count++;
+			this_cpu_ptr(smca_banks)[bank].hwid = s_hwid;
+			this_cpu_ptr(smca_banks)[bank].id = low;
+			this_cpu_ptr(smca_banks)[bank].sysfs_id = bank_counts[s_hwid->bank_type]++;
 			break;
 		}
 	}
@@ -600,7 +611,7 @@ out:
 
 bool amd_filter_mce(struct mce *m)
 {
-	enum smca_bank_types bank_type = smca_get_bank_type(m->bank);
+	enum smca_bank_types bank_type = smca_get_bank_type(m->extcpu, m->bank);
 	struct cpuinfo_x86 *c = &boot_cpu_data;
 
 	/* See Family 17h Models 10h-2Fh Erratum #1114. */
@@ -638,7 +649,7 @@ static void disable_err_thresholding(struct cpuinfo_x86 *c, unsigned int bank)
 	} else if (c->x86 == 0x17 &&
 		   (c->x86_model >= 0x10 && c->x86_model <= 0x2F)) {
 
-		if (smca_get_bank_type(bank) != SMCA_IF)
+		if (smca_get_bank_type(smp_processor_id(), bank) != SMCA_IF)
 			return;
 
 		msrs[0] = MSR_AMD64_SMCA_MCx_MISC(bank);
@@ -706,7 +717,7 @@ bool amd_mce_is_memory_error(struct mce *m)
 	u8 xec = (m->status >> 16) & 0x1f;
 
 	if (mce_flags.smca)
-		return smca_get_bank_type(m->bank) == SMCA_UMC && xec == 0x0;
+		return smca_get_bank_type(m->extcpu, m->bank) == SMCA_UMC && xec == 0x0;
 
 	return m->bank == 4 && xec == 0x8;
 }
@@ -1022,7 +1033,7 @@ static struct kobj_type threshold_ktype = {
 	.release		= threshold_block_release,
 };
 
-static const char *get_name(unsigned int bank, struct threshold_block *b)
+static const char *get_name(unsigned int cpu, unsigned int bank, struct threshold_block *b)
 {
 	enum smca_bank_types bank_type;
 
@@ -1033,7 +1044,7 @@ static const char *get_name(unsigned int bank, struct threshold_block *b)
 		return th_names[bank];
 	}
 
-	bank_type = smca_get_bank_type(bank);
+	bank_type = smca_get_bank_type(cpu, bank);
 	if (bank_type >= N_SMCA_BANK_TYPES)
 		return NULL;
 
@@ -1043,12 +1054,12 @@ static const char *get_name(unsigned int bank, struct threshold_block *b)
 		return NULL;
 	}
 
-	if (smca_banks[bank].hwid->count == 1)
+	if (per_cpu(smca_bank_counts, cpu)[bank_type] == 1)
 		return smca_get_name(bank_type);
 
 	snprintf(buf_mcatype, MAX_MCATYPE_NAME_LEN,
-		 "%s_%x", smca_get_name(bank_type),
-			  smca_banks[bank].sysfs_id);
+		 "%s_%u", smca_get_name(bank_type),
+			  per_cpu(smca_banks, cpu)[bank].sysfs_id);
 	return buf_mcatype;
 }
 
@@ -1104,7 +1115,7 @@ static int allocate_threshold_blocks(unsigned int cpu, struct threshold_bank *tb
 	else
 		tb->blocks = b;
 
-	err = kobject_init_and_add(&b->kobj, &threshold_ktype, tb->kobj, get_name(bank, b));
+	err = kobject_init_and_add(&b->kobj, &threshold_ktype, tb->kobj, get_name(cpu, bank, b));
 	if (err)
 		goto out_free;
 recurse:
@@ -1159,7 +1170,7 @@ static int threshold_create_bank(struct threshold_bank **bp, unsigned int cpu,
 	struct device *dev = this_cpu_read(mce_device);
 	struct amd_northbridge *nb = NULL;
 	struct threshold_bank *b = NULL;
-	const char *name = get_name(bank, NULL);
+	const char *name = get_name(cpu, bank, NULL);
 	int err = 0;
 
 	if (!dev)
diff --git a/drivers/edac/mce_amd.c b/drivers/edac/mce_amd.c
index cfd3f7a..cc5c63f 100644
--- a/drivers/edac/mce_amd.c
+++ b/drivers/edac/mce_amd.c
@@ -1166,20 +1166,13 @@ static void decode_mc6_mce(struct mce *m)
 /* Decode errors according to Scalable MCA specification */
 static void decode_smca_error(struct mce *m)
 {
-	struct smca_hwid *hwid;
-	enum smca_bank_types bank_type;
+	enum smca_bank_types bank_type = smca_get_bank_type(m->extcpu, m->bank);
 	const char *ip_name;
 	u8 xec = XEC(m->status, xec_mask);
 
-	if (m->bank >= ARRAY_SIZE(smca_banks))
+	if (bank_type >= N_SMCA_BANK_TYPES)
 		return;
 
-	hwid = smca_banks[m->bank].hwid;
-	if (!hwid)
-		return;
-
-	bank_type = hwid->bank_type;
-
 	if (bank_type == SMCA_RESERVED) {
 		pr_emerg(HW_ERR "Bank %d is reserved.\n", m->bank);
 		return;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 08133de..75dad02 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -2647,7 +2647,7 @@ static int amdgpu_bad_page_notifier(struct notifier_block *nb,
 	 * and error occurred in DramECC (Extended error code = 0) then only
 	 * process the error, else bail out.
 	 */
-	if (!m || !((smca_get_bank_type(m->bank) == SMCA_UMC_V2) &&
+	if (!m || !((smca_get_bank_type(m->extcpu, m->bank) == SMCA_UMC_V2) &&
 		    (XEC(m->status, 0x3f) == 0x0)))
 		return NOTIFY_DONE;
 

[tip: ras/core] x86/MCE/AMD, EDAC/mce_amd: Add new SMCA bank types

[tip-bot2 for Muralidhara M K]

The following commit has been merged into the ras/core branch of tip:

Commit-ID:     94a311ce248e0b53c76e110fd00511af47b72ffb
Gitweb:        https://git.kernel.org/tip/94a311ce248e0b53c76e110fd00511af47b72ffb
Author:        Muralidhara M K <muralimk@amd.com>
AuthorDate:    Wed, 26 May 2021 22:16:01 +05:30
Committer:     Borislav Petkov <bp@suse.de>
CommitterDate: Thu, 27 May 2021 20:08:14 +02:00

x86/MCE/AMD, EDAC/mce_amd: Add new SMCA bank types

Add the (HWID, MCATYPE) tuples and names for new SMCA bank types.

Also, add their respective error descriptions to the MCE decoding module
edac_mce_amd. Also while at it, optimize the string names for some SMCA
banks.

 [ bp: Drop repeated comments, explain why UMC_V2 is a separate entry. ]

Signed-off-by: Muralidhara M K <muralimk@amd.com>
Signed-off-by: Naveen Krishna Chatradhi  <nchatrad@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Yazen Ghannam <yazen.ghannam@amd.com>
Link: https://lkml.kernel.org/r/20210526164601.66228-1-nchatrad@amd.com
---
 arch/x86/include/asm/mce.h    | 13 ++++--
 arch/x86/kernel/cpu/mce/amd.c | 55 ++++++++++++++++-----------
 drivers/edac/mce_amd.c        | 70 ++++++++++++++++++++++++++++++++++-
 3 files changed, 113 insertions(+), 25 deletions(-)

diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h
index ddfb3ca..0607ec4 100644
--- a/arch/x86/include/asm/mce.h
+++ b/arch/x86/include/asm/mce.h
@@ -305,7 +305,7 @@ extern void apei_mce_report_mem_error(int corrected,
 /* These may be used by multiple smca_hwid_mcatypes */
 enum smca_bank_types {
 	SMCA_LS = 0,	/* Load Store */
-	SMCA_LS_V2,	/* Load Store */
+	SMCA_LS_V2,
 	SMCA_IF,	/* Instruction Fetch */
 	SMCA_L2_CACHE,	/* L2 Cache */
 	SMCA_DE,	/* Decoder Unit */
@@ -314,17 +314,22 @@ enum smca_bank_types {
 	SMCA_FP,	/* Floating Point */
 	SMCA_L3_CACHE,	/* L3 Cache */
 	SMCA_CS,	/* Coherent Slave */
-	SMCA_CS_V2,	/* Coherent Slave */
+	SMCA_CS_V2,
 	SMCA_PIE,	/* Power, Interrupts, etc. */
 	SMCA_UMC,	/* Unified Memory Controller */
+	SMCA_UMC_V2,
 	SMCA_PB,	/* Parameter Block */
 	SMCA_PSP,	/* Platform Security Processor */
-	SMCA_PSP_V2,	/* Platform Security Processor */
+	SMCA_PSP_V2,
 	SMCA_SMU,	/* System Management Unit */
-	SMCA_SMU_V2,	/* System Management Unit */
+	SMCA_SMU_V2,
 	SMCA_MP5,	/* Microprocessor 5 Unit */
 	SMCA_NBIO,	/* Northbridge IO Unit */
 	SMCA_PCIE,	/* PCI Express Unit */
+	SMCA_PCIE_V2,
+	SMCA_XGMI_PCS,	/* xGMI PCS Unit */
+	SMCA_XGMI_PHY,	/* xGMI PHY Unit */
+	SMCA_WAFL_PHY,	/* WAFL PHY Unit */
 	N_SMCA_BANK_TYPES
 };
 
diff --git a/arch/x86/kernel/cpu/mce/amd.c b/arch/x86/kernel/cpu/mce/amd.c
index e486f96..08831ac 100644
--- a/arch/x86/kernel/cpu/mce/amd.c
+++ b/arch/x86/kernel/cpu/mce/amd.c
@@ -77,27 +77,29 @@ struct smca_bank_name {
 };
 
 static struct smca_bank_name smca_names[] = {
-	[SMCA_LS]	= { "load_store",	"Load Store Unit" },
-	[SMCA_LS_V2]	= { "load_store",	"Load Store Unit" },
-	[SMCA_IF]	= { "insn_fetch",	"Instruction Fetch Unit" },
-	[SMCA_L2_CACHE]	= { "l2_cache",		"L2 Cache" },
-	[SMCA_DE]	= { "decode_unit",	"Decode Unit" },
-	[SMCA_RESERVED]	= { "reserved",		"Reserved" },
-	[SMCA_EX]	= { "execution_unit",	"Execution Unit" },
-	[SMCA_FP]	= { "floating_point",	"Floating Point Unit" },
-	[SMCA_L3_CACHE]	= { "l3_cache",		"L3 Cache" },
-	[SMCA_CS]	= { "coherent_slave",	"Coherent Slave" },
-	[SMCA_CS_V2]	= { "coherent_slave",	"Coherent Slave" },
-	[SMCA_PIE]	= { "pie",		"Power, Interrupts, etc." },
-	[SMCA_UMC]	= { "umc",		"Unified Memory Controller" },
-	[SMCA_PB]	= { "param_block",	"Parameter Block" },
-	[SMCA_PSP]	= { "psp",		"Platform Security Processor" },
-	[SMCA_PSP_V2]	= { "psp",		"Platform Security Processor" },
-	[SMCA_SMU]	= { "smu",		"System Management Unit" },
-	[SMCA_SMU_V2]	= { "smu",		"System Management Unit" },
-	[SMCA_MP5]	= { "mp5",		"Microprocessor 5 Unit" },
-	[SMCA_NBIO]	= { "nbio",		"Northbridge IO Unit" },
-	[SMCA_PCIE]	= { "pcie",		"PCI Express Unit" },
+	[SMCA_LS ... SMCA_LS_V2]	= { "load_store",	"Load Store Unit" },
+	[SMCA_IF]			= { "insn_fetch",	"Instruction Fetch Unit" },
+	[SMCA_L2_CACHE]			= { "l2_cache",		"L2 Cache" },
+	[SMCA_DE]			= { "decode_unit",	"Decode Unit" },
+	[SMCA_RESERVED]			= { "reserved",		"Reserved" },
+	[SMCA_EX]			= { "execution_unit",	"Execution Unit" },
+	[SMCA_FP]			= { "floating_point",	"Floating Point Unit" },
+	[SMCA_L3_CACHE]			= { "l3_cache",		"L3 Cache" },
+	[SMCA_CS ... SMCA_CS_V2]	= { "coherent_slave",	"Coherent Slave" },
+	[SMCA_PIE]			= { "pie",		"Power, Interrupts, etc." },
+
+	/* UMC v2 is separate because both of them can exist in a single system. */
+	[SMCA_UMC]			= { "umc",		"Unified Memory Controller" },
+	[SMCA_UMC_V2]			= { "umc_v2",		"Unified Memory Controller v2" },
+	[SMCA_PB]			= { "param_block",	"Parameter Block" },
+	[SMCA_PSP ... SMCA_PSP_V2]	= { "psp",		"Platform Security Processor" },
+	[SMCA_SMU ... SMCA_SMU_V2]	= { "smu",		"System Management Unit" },
+	[SMCA_MP5]			= { "mp5",		"Microprocessor 5 Unit" },
+	[SMCA_NBIO]			= { "nbio",		"Northbridge IO Unit" },
+	[SMCA_PCIE ... SMCA_PCIE_V2]	= { "pcie",		"PCI Express Unit" },
+	[SMCA_XGMI_PCS]			= { "xgmi_pcs",		"Ext Global Memory Interconnect PCS Unit" },
+	[SMCA_XGMI_PHY]			= { "xgmi_phy",		"Ext Global Memory Interconnect PHY Unit" },
+	[SMCA_WAFL_PHY]			= { "wafl_phy",		"WAFL PHY Unit" },
 };
 
 static const char *smca_get_name(enum smca_bank_types t)
@@ -155,6 +157,7 @@ static struct smca_hwid smca_hwid_mcatypes[] = {
 
 	/* Unified Memory Controller MCA type */
 	{ SMCA_UMC,	 HWID_MCATYPE(0x96, 0x0)	},
+	{ SMCA_UMC_V2,	 HWID_MCATYPE(0x96, 0x1)	},
 
 	/* Parameter Block MCA type */
 	{ SMCA_PB,	 HWID_MCATYPE(0x05, 0x0)	},
@@ -175,6 +178,16 @@ static struct smca_hwid smca_hwid_mcatypes[] = {
 
 	/* PCI Express Unit MCA type */
 	{ SMCA_PCIE,	 HWID_MCATYPE(0x46, 0x0)	},
+	{ SMCA_PCIE_V2,	 HWID_MCATYPE(0x46, 0x1)	},
+
+	/* xGMI PCS MCA type */
+	{ SMCA_XGMI_PCS, HWID_MCATYPE(0x50, 0x0)	},
+
+	/* xGMI PHY MCA type */
+	{ SMCA_XGMI_PHY, HWID_MCATYPE(0x259, 0x0)	},
+
+	/* WAFL PHY MCA type */
+	{ SMCA_WAFL_PHY, HWID_MCATYPE(0x267, 0x0)	},
 };
 
 struct smca_bank smca_banks[MAX_NR_BANKS];
diff --git a/drivers/edac/mce_amd.c b/drivers/edac/mce_amd.c
index 5dd905a..43ba0f9 100644
--- a/drivers/edac/mce_amd.c
+++ b/drivers/edac/mce_amd.c
@@ -323,6 +323,21 @@ static const char * const smca_umc_mce_desc[] = {
 	"AES SRAM ECC error",
 };
 
+static const char * const smca_umc2_mce_desc[] = {
+	"DRAM ECC error",
+	"Data poison error",
+	"SDP parity error",
+	"Reserved",
+	"Address/Command parity error",
+	"Write data parity error",
+	"DCQ SRAM ECC error",
+	"Reserved",
+	"Read data parity error",
+	"Rdb SRAM ECC error",
+	"RdRsp SRAM ECC error",
+	"LM32 MP errors",
+};
+
 static const char * const smca_pb_mce_desc[] = {
 	"An ECC error in the Parameter Block RAM array",
 };
@@ -400,6 +415,56 @@ static const char * const smca_pcie_mce_desc[] = {
 	"CCIX Non-okay write response with data error",
 };
 
+static const char * const smca_pcie2_mce_desc[] = {
+	"SDP Parity Error logging",
+};
+
+static const char * const smca_xgmipcs_mce_desc[] = {
+	"Data Loss Error",
+	"Training Error",
+	"Flow Control Acknowledge Error",
+	"Rx Fifo Underflow Error",
+	"Rx Fifo Overflow Error",
+	"CRC Error",
+	"BER Exceeded Error",
+	"Tx Vcid Data Error",
+	"Replay Buffer Parity Error",
+	"Data Parity Error",
+	"Replay Fifo Overflow Error",
+	"Replay FIfo Underflow Error",
+	"Elastic Fifo Overflow Error",
+	"Deskew Error",
+	"Flow Control CRC Error",
+	"Data Startup Limit Error",
+	"FC Init Timeout Error",
+	"Recovery Timeout Error",
+	"Ready Serial Timeout Error",
+	"Ready Serial Attempt Error",
+	"Recovery Attempt Error",
+	"Recovery Relock Attempt Error",
+	"Replay Attempt Error",
+	"Sync Header Error",
+	"Tx Replay Timeout Error",
+	"Rx Replay Timeout Error",
+	"LinkSub Tx Timeout Error",
+	"LinkSub Rx Timeout Error",
+	"Rx CMD Pocket Error",
+};
+
+static const char * const smca_xgmiphy_mce_desc[] = {
+	"RAM ECC Error",
+	"ARC instruction buffer parity error",
+	"ARC data buffer parity error",
+	"PHY APB error",
+};
+
+static const char * const smca_waflphy_mce_desc[] = {
+	"RAM ECC Error",
+	"ARC instruction buffer parity error",
+	"ARC data buffer parity error",
+	"PHY APB error",
+};
+
 struct smca_mce_desc {
 	const char * const *descs;
 	unsigned int num_descs;
@@ -418,6 +483,7 @@ static struct smca_mce_desc smca_mce_descs[] = {
 	[SMCA_CS_V2]	= { smca_cs2_mce_desc,	ARRAY_SIZE(smca_cs2_mce_desc)	},
 	[SMCA_PIE]	= { smca_pie_mce_desc,	ARRAY_SIZE(smca_pie_mce_desc)	},
 	[SMCA_UMC]	= { smca_umc_mce_desc,	ARRAY_SIZE(smca_umc_mce_desc)	},
+	[SMCA_UMC_V2]	= { smca_umc2_mce_desc,	ARRAY_SIZE(smca_umc2_mce_desc)	},
 	[SMCA_PB]	= { smca_pb_mce_desc,	ARRAY_SIZE(smca_pb_mce_desc)	},
 	[SMCA_PSP]	= { smca_psp_mce_desc,	ARRAY_SIZE(smca_psp_mce_desc)	},
 	[SMCA_PSP_V2]	= { smca_psp2_mce_desc,	ARRAY_SIZE(smca_psp2_mce_desc)	},
@@ -426,6 +492,10 @@ static struct smca_mce_desc smca_mce_descs[] = {
 	[SMCA_MP5]	= { smca_mp5_mce_desc,	ARRAY_SIZE(smca_mp5_mce_desc)	},
 	[SMCA_NBIO]	= { smca_nbio_mce_desc,	ARRAY_SIZE(smca_nbio_mce_desc)	},
 	[SMCA_PCIE]	= { smca_pcie_mce_desc,	ARRAY_SIZE(smca_pcie_mce_desc)	},
+	[SMCA_PCIE_V2]	= { smca_pcie2_mce_desc,   ARRAY_SIZE(smca_pcie2_mce_desc)	},
+	[SMCA_XGMI_PCS]	= { smca_xgmipcs_mce_desc, ARRAY_SIZE(smca_xgmipcs_mce_desc)	},
+	[SMCA_XGMI_PHY]	= { smca_xgmiphy_mce_desc, ARRAY_SIZE(smca_xgmiphy_mce_desc)	},
+	[SMCA_WAFL_PHY]	= { smca_waflphy_mce_desc, ARRAY_SIZE(smca_waflphy_mce_desc)	},
 };
 
 static bool f12h_mc0_mce(u16 ec, u8 xec)