[PATCH 0/2] rasdaemon: Add new AMD SMCA error decoding

[PATCH 0/2] rasdaemon: Add new AMD SMCA error decoding

[Ghannam, Yazen]

From: Yazen Ghannam <yazen.ghannam@amd.com>

Hi Mauro,

This patchset adds new error decoding for recently released AMD Scalable
MCA systems. Also, the CPU family check is changed to a feature check.
This is because the Scalable MCA decoding should function the same on
any AMD system with this feature.

Also, this set was written by Brian Woods at AMD, but he has since left.
So I'm submitting on his behalf. Please let me know if I should re-do
the Signed-off-by's in another way.

Thanks,
Yazen

Brian Woods (2):
  rasdaemon: rename CPU_NAPLES cputype
  rasdaemon: add support for new AMD SMCA bank types

 mce-amd-smca.c    | 112 ++++++++++++++++++++++++++++++++++++++++++++++
 ras-mce-handler.c |  10 +++--
 ras-mce-handler.h |   2 +-
 3 files changed, 119 insertions(+), 5 deletions(-)

-- 
2.17.1

[PATCH 1/2] rasdaemon: rename CPU_NAPLES cputype

[Ghannam, Yazen]

From: Brian Woods <brian.woods@amd.com>

Change CPU_NAPLES to CPU_AMD_SMCA to reflect that it isn't just NAPLES
that is supported, but AMD's Scalable Machine Check Architecture (SMCA).

  [ Yazen: change family check to feature check, and change CPU name. ]

Signed-off-by: Brian Woods <brian.woods@amd.com>
Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Cc: Chandu-babu Namburu <chandu@amd.com>
---
 ras-mce-handler.c | 10 ++++++----
 ras-mce-handler.h |  2 +-
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/ras-mce-handler.c b/ras-mce-handler.c
index fd3ef3b..016acae 100644
--- a/ras-mce-handler.c
+++ b/ras-mce-handler.c
@@ -55,7 +55,7 @@ static char *cputype_name[] = {
 	[CPU_KNIGHTS_LANDING] = "Knights Landing",
 	[CPU_KNIGHTS_MILL] = "Knights Mill",
 	[CPU_SKYLAKE_XEON] = "Skylake server",
-	[CPU_NAPLES] = "AMD Family 17h Zen1",
+	[CPU_AMD_SMCA] = "AMD Scalable MCA",
 	[CPU_DHYANA] = "Hygon Family 18h Moksha"
 };
 
@@ -192,8 +192,10 @@ static int detect_cpu(struct ras_events *ras)
 	if (!strcmp(mce->vendor, "AuthenticAMD")) {
 		if (mce->family == 15)
 			mce->cputype = CPU_K8;
-		if (mce->family == 23)
-			mce->cputype = CPU_NAPLES;
+		if (strstr(mce->processor_flags, "smca")) {
+			mce->cputype = CPU_AMD_SMCA;
+			goto ret;
+		}
 		if (mce->family > 23) {
 			log(ALL, LOG_INFO,
 			    "Can't parse MCE for this AMD CPU yet %d\n",
@@ -441,7 +443,7 @@ int ras_mce_event_handler(struct trace_seq *s,
 	case CPU_K8:
 		rc = parse_amd_k8_event(ras, &e);
 		break;
-	case CPU_NAPLES:
+	case CPU_AMD_SMCA:
 	case CPU_DHYANA:
 		rc = parse_amd_smca_event(ras, &e);
 		break;
diff --git a/ras-mce-handler.h b/ras-mce-handler.h
index 94395eb..1ef7ef2 100644
--- a/ras-mce-handler.h
+++ b/ras-mce-handler.h
@@ -50,7 +50,7 @@ enum cputype {
 	CPU_KNIGHTS_LANDING,
 	CPU_KNIGHTS_MILL,
 	CPU_SKYLAKE_XEON,
-	CPU_NAPLES,
+	CPU_AMD_SMCA,
 	CPU_DHYANA,
 };
 
-- 
2.17.1

[PATCH 2/2] rasdaemon: add support for new AMD SMCA bank types

[Ghannam, Yazen]

From: Brian Woods <brian.woods@amd.com>

Going forward, the Scalable Machine Check Architecture (SMCA) has some
updated and additional bank types which show up in Zen2.  The differing
bank types include: CS_V2, PSP_V2, SMU_V2, MP5, NBIO, and PCIE.  The V2
bank types replace the original bank types but have unique HWID/MCAtype
IDs from the originals so there's no conflicts between different
versions or other bank types.  All of the differing bank types have new
MCE descriptions which have been added as well.

Signed-off-by: Brian Woods <brian.woods@amd.com>
Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Cc: Chandu-babu Namburu <chandu@amd.com>
---
 mce-amd-smca.c | 112 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 112 insertions(+)

diff --git a/mce-amd-smca.c b/mce-amd-smca.c
index 6c3e8a5..114e786 100644
--- a/mce-amd-smca.c
+++ b/mce-amd-smca.c
@@ -49,11 +49,17 @@ enum smca_bank_types {
 	SMCA_FP,        /* Floating Point */
 	SMCA_L3_CACHE,  /* L3 Cache */
 	SMCA_CS,        /* Coherent Slave */
+	SMCA_CS_V2,     /* Coherent Slave V2 */
 	SMCA_PIE,       /* Power, Interrupts, etc. */
 	SMCA_UMC,       /* Unified Memory Controller */
 	SMCA_PB,        /* Parameter Block */
 	SMCA_PSP,       /* Platform Security Processor */
+	SMCA_PSP_V2,    /* Platform Security Processor V2 */
 	SMCA_SMU,       /* System Management Unit */
+	SMCA_SMU_V2,    /* System Management Unit V2 */
+	SMCA_MP5,	/* Microprocessor 5 Unit */
+	SMCA_NBIO,	/* Northbridge IO Unit */
+	SMCA_PCIE,	/* PCI Express Unit */
 	N_SMCA_BANK_TYPES
 };
 
@@ -165,6 +171,23 @@ static const char * const smca_cs_mce_desc[] = {
 	"Atomic request parity",
 	"ECC error on probe filter access",
 };
+/* Coherent Slave Unit V2 */
+static const char * const smca_cs2_mce_desc[] = {
+	"Illegal Request",
+	"Address Violation",
+	"Security Violation",
+	"Illegal Response",
+	"Unexpected Response",
+	"Request or Probe Parity Error",
+	"Read Response Parity Error",
+	"Atomic Request Parity Error",
+	"SDP read response had no match in the CS queue",
+	"Probe Filter Protocol Error",
+	"Probe Filter ECC Error",
+	"SDP read response had an unexpected RETRY error",
+	"Counter overflow error",
+	"Counter underflow error",
+};
 /* Power, Interrupt, etc.. */
 static const char * const smca_pie_mce_desc[] = {
 	"HW assert",
@@ -189,10 +212,75 @@ static const char * const smca_pb_mce_desc[] = {
 static const char * const smca_psp_mce_desc[] = {
 	"PSP RAM ECC or parity error",
 };
+/* Platform Security Processor V2 */
+static const char * const smca_psp2_mce_desc[] = {
+	"High SRAM ECC or parity error",
+	"Low SRAM ECC or parity error",
+	"Instruction Cache Bank 0 ECC or parity error",
+	"Instruction Cache Bank 1 ECC or parity error",
+	"Instruction Tag Ram 0 parity error",
+	"Instruction Tag Ram 1 parity error",
+	"Data Cache Bank 0 ECC or parity error",
+	"Data Cache Bank 1 ECC or parity error",
+	"Data Cache Bank 2 ECC or parity error",
+	"Data Cache Bank 3 ECC or parity error",
+	"Data Tag Bank 0 parity error",
+	"Data Tag Bank 1 parity error",
+	"Data Tag Bank 2 parity error",
+	"Data Tag Bank 3 parity error",
+	"Dirty Data Ram parity error",
+	"TLB Bank 0 parity error",
+	"TLB Bank 1 parity error",
+	"System Hub Read Buffer ECC or parity error",
+};
 /* System Management Unit */
 static const char * const smca_smu_mce_desc[] = {
 	"SMU RAM ECC or parity error",
 };
+/* System Management Unit V2 */
+static const char * const smca_smu2_mce_desc[] = {
+	"High SRAM ECC or parity error",
+	"Low SRAM 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",
+};
+/* Microprocessor 5 Unit */
+static const char * const smca_mp5_mce_desc[] = {
+	"High SRAM ECC or parity error",
+	"Low SRAM 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",
+};
+/* Northbridge IO Unit */
+static const char * const smca_nbio_mce_desc[] = {
+	"ECC or Parity error",
+	"PCIE error",
+	"SDP ErrEvent error",
+	"SDP Egress Poison Error",
+	"IOHC Internal Poison Error",
+};
+/* PCI Express Unit */
+static const char * const smca_pcie_mce_desc[] = {
+	"CCIX PER Message logging",
+	"CCIX Read Response with Status: Non-Data Error",
+	"CCIX Write Response with Status: Non-Data Error",
+	"CCIX Read Response with Status: Data Error",
+	"CCIX Non-okay write response with data error",
+};
+
 
 struct smca_mce_desc {
 	const char * const *descs;
@@ -208,11 +296,17 @@ static struct smca_mce_desc smca_mce_descs[] = {
 	[SMCA_FP]       = { smca_fp_mce_desc,   ARRAY_SIZE(smca_fp_mce_desc)  },
 	[SMCA_L3_CACHE] = { smca_l3_mce_desc,   ARRAY_SIZE(smca_l3_mce_desc)  },
 	[SMCA_CS]       = { smca_cs_mce_desc,   ARRAY_SIZE(smca_cs_mce_desc)  },
+	[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_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)},
 	[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_NBIO]     = { smca_nbio_mce_desc, ARRAY_SIZE(smca_nbio_mce_desc)},
+	[SMCA_PCIE]     = { smca_pcie_mce_desc, ARRAY_SIZE(smca_pcie_mce_desc)},
 };
 
 struct smca_hwid {
@@ -235,6 +329,7 @@ static struct smca_hwid smca_hwid_mcatypes[] = {
 
 	/* Data Fabric MCA types */
 	{ SMCA_CS,       0x0000002E },
+	{ SMCA_CS_V2,    0x0002002E },
 	{ SMCA_PIE,      0x0001002E },
 
 	/* Unified Memory Controller MCA type */
@@ -245,9 +340,20 @@ static struct smca_hwid smca_hwid_mcatypes[] = {
 
 	/* Platform Security Processor MCA type */
 	{ SMCA_PSP,      0x000000FF },
+	{ SMCA_PSP_V2,   0x000100FF },
 
 	/* System Management Unit MCA type */
 	{ SMCA_SMU,      0x00000001 },
+	{ SMCA_SMU_V2,   0x00010001 },
+
+	/* Microprocessor 5 Unit MCA type */
+	{ SMCA_MP5,      0x00020001 },
+
+	/* Northbridge IO Unit MCA type */
+	{ SMCA_NBIO,     0x00000018 },
+
+	/* PCI Express Unit MCA type */
+	{ SMCA_PCIE,     0x00000046 },
 };
 
 struct smca_bank_name {
@@ -264,11 +370,17 @@ static struct smca_bank_name smca_names[] = {
 	[SMCA_FP]       = { "Floating Point Unit" },
 	[SMCA_L3_CACHE] = { "L3 Cache" },
 	[SMCA_CS]       = { "Coherent Slave" },
+	[SMCA_CS_V2]    = { "Coherent Slave" },
 	[SMCA_PIE]      = { "Power, Interrupts, etc." },
 	[SMCA_UMC]      = { "Unified Memory Controller" },
 	[SMCA_PB]       = { "Parameter Block" },
 	[SMCA_PSP]      = { "Platform Security Processor" },
+	[SMCA_PSP_V2]   = { "Platform Security Processor" },
 	[SMCA_SMU]      = { "System Management Unit" },
+	[SMCA_SMU_V2]   = { "System Management Unit" },
+	[SMCA_MP5]	= { "Microprocessor 5 Unit" },
+	[SMCA_NBIO]     = { "Northbridge IO Unit" },
+	[SMCA_PCIE]     = { "PCI Express Unit" },
 };
 
 static void amd_decode_errcode(struct mce_event *e)
-- 
2.17.1