| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20241120093745.1847-5-shiju.jose@huawei.com>
Date: Wed, 20 Nov 2024 09:37:43 +0000
From: <shiju.jose@...wei.com>
To: <dave.jiang@...el.com>, <dan.j.williams@...el.com>,
<jonathan.cameron@...wei.com>, <alison.schofield@...el.com>,
<nifan.cxl@...il.com>, <vishal.l.verma@...el.com>, <ira.weiny@...el.com>,
<dave@...olabs.net>, <linux-cxl@...r.kernel.org>
CC: <linux-kernel@...r.kernel.org>, <linuxarm@...wei.com>,
<tanxiaofei@...wei.com>, <prime.zeng@...ilicon.com>, <shiju.jose@...wei.com>
Subject: [PATCH v4 4/6] cxl/events: Update DRAM Event Record to CXL spec rev 3.1
From: Shiju Jose <shiju.jose@...wei.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.
Example trace print of cxl_dram trace event,
cxl_dram: memdev=mem0 host=0000:0f:00.0 serial=3 log=Informational : \
time=76068483335 uuid=601dcbb3-9c06-4eab-b8af-4e9bfb5c9624 len=128 \
flags='0x1' handle=1 related_handle=0 maint_op_class=1 \
maint_op_sub_class=3 : dpa=0x18680 dpa_flags=0x0 descriptor=0x3 \
type=0x3 sub_type=0x5 transaction_type=0x5 channel=3 rank=17 \
nibble_mask=0x3b00b2 bank_group=7 bank=11 row=2 column=77 \
cor_mask=21 00 00 00 00 00 00 00 2c 00 00 00 00 00 00 00 37 00 00 00 \
00 00 00 00 42 00 00 00 00 00 00 00 validity_flags=0x3ff \
comp_id=01 74 c5 08 9a 1a 0b fc d2 7e 2f 31 9b 3c 81 4d \
pldm_entity_id=74 c5 08 9a 1a 0b pldm_resource_id=0x00 \
hpa=0xffffffffffffffff region= \
region_uuid=00000000-0000-0000-0000-000000000000 \
sub_channel=5 cme_threshold_ev_flags=0x3 cvme_count=0x94
The number of decoded strings in TP_printk() caused a parsing error when
libtraceevent in userspace parsed the CXL dram trace event for rasdaemon.
It was found that long decoded strings of field values in the TP_printk()
caused the issue. As a solution, decoding of field values in the TP_printk()
were removed to accommodate the new fields.
Decoding of all these fields is added in the userspace tool rasdaemon.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@...wei.com>
Reviewed-by: Davidlohr Bueso <dave@...olabs.net>
Signed-off-by: Shiju Jose <shiju.jose@...wei.com>
---
drivers/cxl/core/trace.h | 82 +++++++++++++++++++++++++++-------------
include/cxl/event.h | 9 ++++-
2 files changed, 63 insertions(+), 28 deletions(-)
diff --git a/drivers/cxl/core/trace.h b/drivers/cxl/core/trace.h
index 77055d66b56e..f4f3d49166cd 100644
--- a/drivers/cxl/core/trace.h
+++ b/drivers/cxl/core/trace.h
@@ -435,7 +435,7 @@ 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
@@ -445,11 +445,17 @@ 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)
@@ -460,15 +466,21 @@ 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_COMPONENT_ID_FORMAT, "COMPONENT PLDM FORMAT" }, \
+ { CXL_DER_VALID_SUB_CHANNEL, "SUB CHANNEL" } \
)
TRACE_EVENT(cxl_dram,
@@ -484,6 +496,7 @@ 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)
@@ -498,6 +511,10 @@ TRACE_EVENT(cxl_dram,
__field(u8, bank) /* Out of order to pack trace record */
__field(u8, dpa_flags) /* Out of order to pack trace record */
__string(region_name, cxlr ? dev_name(&cxlr->dev) : "")
+ __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)
),
TP_fast_assign(
@@ -510,6 +527,7 @@ 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;
@@ -529,23 +547,35 @@ TRACE_EVENT(cxl_dram,
__assign_str(region_name);
uuid_copy(&__entry->region_uuid, &uuid_null);
}
+ memcpy(__entry->comp_id, &rec->component_id,
+ CXL_EVENT_GEN_MED_COMP_ID_SIZE);
+ __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);
),
- CXL_EVT_TP_printk("dpa=%llx dpa_flags='%s' descriptor='%s' 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",
- __entry->dpa, show_dpa_flags(__entry->dpa_flags),
- show_event_desc_flags(__entry->descriptor),
- show_dram_mem_event_type(__entry->type),
- show_trans_type(__entry->transaction_type),
+ CXL_EVT_TP_printk("dpa=0x%llx dpa_flags=0x%x " \
+ "descriptor=0x%x type=0x%x sub_type=0x%x " \
+ "transaction_type=0x%x channel=%u rank=%u " \
+ "nibble_mask=0x%x bank_group=%u bank=%u " \
+ "row=%u column=%u cor_mask=%s validity_flags=0x%x " \
+ "comp_id=%s pldm_entity_id=%s pldm_resource_id=%s " \
+ "hpa=0x%llx region=%s region_uuid=%pUb " \
+ "sub_channel=%u cme_threshold_ev_flags=0x%x cvme_count=0x%x ", \
+ __entry->dpa, __entry->dpa_flags, __entry->descriptor,
+ __entry->type, __entry->sub_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),
- show_dram_valid_flags(__entry->validity_flags),
- __entry->hpa, __get_str(region_name), &__entry->region_uuid
+ __entry->validity_flags,
+ __print_hex(__entry->comp_id, CXL_EVENT_GEN_MED_COMP_ID_SIZE),
+ show_pldm_entity_id(__entry->validity_flags, CXL_DER_VALID_COMPONENT,
+ CXL_DER_VALID_COMPONENT_ID_FORMAT, __entry->comp_id),
+ show_pldm_resource_id(__entry->validity_flags, CXL_DER_VALID_COMPONENT,
+ CXL_DER_VALID_COMPONENT_ID_FORMAT, __entry->comp_id),
+ __entry->hpa, __get_str(region_name), &__entry->region_uuid,
+ __entry->sub_channel, __entry->cme_threshold_ev_flags, __entry->cvme_count
)
);
diff --git a/include/cxl/event.h b/include/cxl/event.h
index 2b07adf39010..dd85aa9beddf 100644
--- a/include/cxl/event.h
+++ b/include/cxl/event.h
@@ -60,7 +60,7 @@ struct cxl_event_gen_media {
/*
* DRAM Event Record - DER
- * CXL rev 3.0 section 8.2.9.2.1.2; Table 3-44
+ * CXL rev 3.1 section 8.2.9.2.1.2; Table 3-46
*/
#define CXL_EVENT_DER_CORRECTION_MASK_SIZE 0x20
struct cxl_event_dram {
@@ -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;
/*
--
2.43.0
Powered by blists - more mailing lists