[PATCH v2 4/6] cxl/events: Update DRAM Event Record to CXL spec rev 3.1

shiju.jose@huawei.com posted 6 patches 1 month ago
There is a newer version of this series
[PATCH v2 4/6] cxl/events: Update DRAM Event Record to CXL spec rev 3.1
Posted by shiju.jose@huawei.com 1 month ago
From: Shiju Jose <shiju.jose@huawei.com>

CXL spec 3.1 section 8.2.9.2.1.2 Table 8-46, DRAM Event Record has updated
with following new fields and new types for Memory Event Type, Transaction
Type and Validity Flags fields.
1. Component Identifier
2. Sub-channel
3. Advanced Programmable Corrected Memory Error Threshold Event Flags
4. Corrected Memory Error Count at Event
5. Memory Event Sub-Type

Update DRAM events record and DRAM trace event for the above spec
changes. The new fields are inserted in logical places.
Includes trivial consistency of white space improvements.

Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Shiju Jose <shiju.jose@huawei.com>
---
 drivers/cxl/core/trace.h | 67 ++++++++++++++++++++++++++++------------
 include/cxl/event.h      |  7 ++++-
 2 files changed, 54 insertions(+), 20 deletions(-)

diff --git a/drivers/cxl/core/trace.h b/drivers/cxl/core/trace.h
index ef70fc37660f..f6142c01d306 100644
--- a/drivers/cxl/core/trace.h
+++ b/drivers/cxl/core/trace.h
@@ -470,7 +470,7 @@ TRACE_EVENT(cxl_general_media,
 /*
  * DRAM Event Record - DER
  *
- * CXL rev 3.0 section 8.2.9.2.1.2; Table 8-44
+ * CXL rev 3.1 section 8.2.9.2.1.2; Table 8-46
  */
 /*
  * DRAM Event Record defines many fields the same as the General Media Event
@@ -480,11 +480,17 @@ TRACE_EVENT(cxl_general_media,
 #define CXL_DER_MEM_EVT_TYPE_SCRUB_MEDIA_ECC_ERROR	0x01
 #define CXL_DER_MEM_EVT_TYPE_INV_ADDR			0x02
 #define CXL_DER_MEM_EVT_TYPE_DATA_PATH_ERROR		0x03
-#define show_dram_mem_event_type(type)  __print_symbolic(type,				\
-	{ CXL_DER_MEM_EVT_TYPE_ECC_ERROR,		"ECC Error" },			\
-	{ CXL_DER_MEM_EVT_TYPE_SCRUB_MEDIA_ECC_ERROR,	"Scrub Media ECC Error" },	\
-	{ CXL_DER_MEM_EVT_TYPE_INV_ADDR,		"Invalid Address" },		\
-	{ CXL_DER_MEM_EVT_TYPE_DATA_PATH_ERROR,		"Data Path Error" }		\
+#define CXL_DER_MEM_EVT_TYPE_TE_STATE_VIOLATION	0x04
+#define CXL_DER_MEM_EVT_TYPE_AP_CME_COUNTER_EXPIRE	0x05
+#define CXL_DER_MEM_EVT_TYPE_CKID_VIOLATION		0x06
+#define show_dram_mem_event_type(type)	__print_symbolic(type,					\
+	{ CXL_DER_MEM_EVT_TYPE_ECC_ERROR,		"ECC Error" },				\
+	{ CXL_DER_MEM_EVT_TYPE_SCRUB_MEDIA_ECC_ERROR,	"Scrub Media ECC Error" },		\
+	{ CXL_DER_MEM_EVT_TYPE_INV_ADDR,		"Invalid Address" },			\
+	{ CXL_DER_MEM_EVT_TYPE_DATA_PATH_ERROR,		"Data Path Error" },			\
+	{ CXL_DER_MEM_EVT_TYPE_TE_STATE_VIOLATION,	"TE State Violation" },			\
+	{ CXL_DER_MEM_EVT_TYPE_AP_CME_COUNTER_EXPIRE,	"Adv Prog CME Counter Expiration" },	\
+	{ CXL_DER_MEM_EVT_TYPE_CKID_VIOLATION,		"CKID Violation" }			\
 )
 
 #define CXL_DER_VALID_CHANNEL				BIT(0)
@@ -495,15 +501,20 @@ TRACE_EVENT(cxl_general_media,
 #define CXL_DER_VALID_ROW				BIT(5)
 #define CXL_DER_VALID_COLUMN				BIT(6)
 #define CXL_DER_VALID_CORRECTION_MASK			BIT(7)
-#define show_dram_valid_flags(flags)	__print_flags(flags, "|",			   \
-	{ CXL_DER_VALID_CHANNEL,			"CHANNEL"		}, \
-	{ CXL_DER_VALID_RANK,				"RANK"			}, \
-	{ CXL_DER_VALID_NIBBLE,				"NIBBLE"		}, \
-	{ CXL_DER_VALID_BANK_GROUP,			"BANK GROUP"		}, \
-	{ CXL_DER_VALID_BANK,				"BANK"			}, \
-	{ CXL_DER_VALID_ROW,				"ROW"			}, \
-	{ CXL_DER_VALID_COLUMN,				"COLUMN"		}, \
-	{ CXL_DER_VALID_CORRECTION_MASK,		"CORRECTION MASK"	}  \
+#define CXL_DER_VALID_COMPONENT				BIT(8)
+#define CXL_DER_VALID_COMPONENT_ID_FORMAT		BIT(9)
+#define CXL_DER_VALID_SUB_CHANNEL			BIT(10)
+#define show_dram_valid_flags(flags)	__print_flags(flags, "|",			\
+	{ CXL_DER_VALID_CHANNEL,			"CHANNEL"		},	\
+	{ CXL_DER_VALID_RANK,				"RANK"			},	\
+	{ CXL_DER_VALID_NIBBLE,				"NIBBLE"		},	\
+	{ CXL_DER_VALID_BANK_GROUP,			"BANK GROUP"		},	\
+	{ CXL_DER_VALID_BANK,				"BANK"			},	\
+	{ CXL_DER_VALID_ROW,				"ROW"			},	\
+	{ CXL_DER_VALID_COLUMN,				"COLUMN"		},	\
+	{ CXL_DER_VALID_CORRECTION_MASK,		"CORRECTION MASK"	},	\
+	{ CXL_DER_VALID_COMPONENT,			"COMPONENT"		},	\
+	{ CXL_DER_VALID_SUB_CHANNEL,			"SUB CHANNEL"		}	\
 )
 
 TRACE_EVENT(cxl_dram,
@@ -519,6 +530,7 @@ TRACE_EVENT(cxl_dram,
 		__field(u64, dpa)
 		__field(u8, descriptor)
 		__field(u8, type)
+		__field(u8, sub_type)
 		__field(u8, transaction_type)
 		__field(u8, channel)
 		__field(u16, validity_flags)
@@ -532,6 +544,10 @@ TRACE_EVENT(cxl_dram,
 		__field(u8, bank_group)	/* Out of order to pack trace record */
 		__field(u8, bank)	/* Out of order to pack trace record */
 		__field(u8, dpa_flags)	/* Out of order to pack trace record */
+		__array(u8, comp_id, CXL_EVENT_GEN_MED_COMP_ID_SIZE)
+		__field(u8, sub_channel)
+		__field(u8, cme_threshold_ev_flags)
+		__field(u32, cvme_count)
 		__string(region_name, cxlr ? dev_name(&cxlr->dev) : "")
 	),
 
@@ -545,6 +561,7 @@ TRACE_EVENT(cxl_dram,
 		__entry->dpa &= CXL_DPA_MASK;
 		__entry->descriptor = rec->media_hdr.descriptor;
 		__entry->type = rec->media_hdr.type;
+		__entry->sub_type = rec->sub_type;
 		__entry->transaction_type = rec->media_hdr.transaction_type;
 		__entry->validity_flags = get_unaligned_le16(rec->media_hdr.validity_flags);
 		__entry->channel = rec->media_hdr.channel;
@@ -556,7 +573,12 @@ TRACE_EVENT(cxl_dram,
 		__entry->column = get_unaligned_le16(rec->column);
 		memcpy(__entry->cor_mask, &rec->correction_mask,
 			CXL_EVENT_DER_CORRECTION_MASK_SIZE);
+		memcpy(__entry->comp_id, &rec->component_id,
+		       CXL_EVENT_GEN_MED_COMP_ID_SIZE);
 		__entry->hpa = hpa;
+		__entry->sub_channel = rec->sub_channel;
+		__entry->cme_threshold_ev_flags = rec->cme_threshold_ev_flags;
+		__entry->cvme_count = get_unaligned_le24(rec->cvme_count);
 		if (cxlr) {
 			__assign_str(region_name);
 			uuid_copy(&__entry->region_uuid, &cxlr->params.uuid);
@@ -566,21 +588,28 @@ TRACE_EVENT(cxl_dram,
 		}
 	),
 
-	CXL_EVT_TP_printk("dpa=%llx dpa_flags='%s' descriptor='%s' type='%s' " \
+	CXL_EVT_TP_printk("dpa=%llx dpa_flags='%s' descriptor='%s' type='%s' sub_type='%s' " \
 		"transaction_type='%s' channel=%u rank=%u nibble_mask=%x " \
 		"bank_group=%u bank=%u row=%u column=%u cor_mask=%s " \
-		"validity_flags='%s' " \
-		"hpa=%llx region=%s region_uuid=%pUb",
+		"comp_id=%s%s validity_flags='%s' " \
+		"hpa=%llx sub_channel=%u cme_threshold_ev_flags='%s' cvme_count=%x " \
+		"region=%s region_uuid=%pUb",
 		__entry->dpa, show_dpa_flags(__entry->dpa_flags),
 		show_event_desc_flags(__entry->descriptor),
 		show_dram_mem_event_type(__entry->type),
+		show_mem_event_sub_type(__entry->sub_type),
 		show_trans_type(__entry->transaction_type),
 		__entry->channel, __entry->rank, __entry->nibble_mask,
 		__entry->bank_group, __entry->bank,
 		__entry->row, __entry->column,
 		__print_hex(__entry->cor_mask, CXL_EVENT_DER_CORRECTION_MASK_SIZE),
+		comp_id_pldm_flags(__entry->comp_id[0]),
+		cxl_print_component_id(__entry->validity_flags, CXL_DER_VALID_COMPONENT,
+				       CXL_DER_VALID_COMPONENT_ID_FORMAT, __entry->comp_id),
 		show_dram_valid_flags(__entry->validity_flags),
-		__entry->hpa, __get_str(region_name), &__entry->region_uuid
+		__entry->hpa, __entry->sub_channel,
+		show_cme_threshold_ev_flags(__entry->cme_threshold_ev_flags),
+		__entry->cvme_count, __get_str(region_name), &__entry->region_uuid
 	)
 );
 
diff --git a/include/cxl/event.h b/include/cxl/event.h
index ea8cd44a52e9..7e98492c85df 100644
--- a/include/cxl/event.h
+++ b/include/cxl/event.h
@@ -71,7 +71,12 @@ struct cxl_event_dram {
 	u8 row[3];
 	u8 column[2];
 	u8 correction_mask[CXL_EVENT_DER_CORRECTION_MASK_SIZE];
-	u8 reserved[0x17];
+	u8 component_id[CXL_EVENT_GEN_MED_COMP_ID_SIZE];
+	u8 sub_channel;
+	u8 cme_threshold_ev_flags;
+	u8 cvme_count[3];
+	u8 sub_type;
+	u8 reserved;
 } __packed;
 
 /*
-- 
2.34.1
Re: [PATCH v2 4/6] cxl/events: Update DRAM Event Record to CXL spec rev 3.1
Posted by Davidlohr Bueso 1 month ago
On Tue, 22 Oct 2024, shiju.jose@huawei.com wrote:\n
>From: Shiju Jose <shiju.jose@huawei.com>
>
>CXL spec 3.1 section 8.2.9.2.1.2 Table 8-46, DRAM Event Record has updated
>with following new fields and new types for Memory Event Type, Transaction
>Type and Validity Flags fields.
>1. Component Identifier
>2. Sub-channel
>3. Advanced Programmable Corrected Memory Error Threshold Event Flags
>4. Corrected Memory Error Count at Event
>5. Memory Event Sub-Type
>
>Update DRAM events record and DRAM trace event for the above spec
>changes. The new fields are inserted in logical places.
>Includes trivial consistency of white space improvements.

Reviewed-by: Davidlohr Bueso <dave@stgolabs.net>