[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20231012230301.58500-4-Smita.KoralahalliChannabasappa@amd.com>
Date: Thu, 12 Oct 2023 23:03:01 +0000
From: Smita Koralahalli <Smita.KoralahalliChannabasappa@....com>
To: <linux-efi@...r.kernel.org>, <linux-cxl@...r.kernel.org>,
<linux-kernel@...r.kernel.org>
CC: Ard Biesheuvel <ardb@...nel.org>,
Alison Schofield <alison.schofield@...el.com>,
Vishal Verma <vishal.l.verma@...el.com>,
"Ira Weiny" <ira.weiny@...el.com>,
Dan Williams <dan.j.williams@...el.com>,
Jonathan Cameron <Jonathan.Cameron@...wei.com>,
Yazen Ghannam <yazen.ghannam@....com>,
Smita Koralahalli <Smita.KoralahalliChannabasappa@....com>
Subject: [PATCH 3/3] efi/cper, cxl: Decode CXL Component Events Memory Module Event Record
Add support for decoding CXL Component Events Memory Module Event Record
as defined in CXL rev 3.0 section 8.2.9.2.1.3.
Signed-off-by: Smita Koralahalli <Smita.KoralahalliChannabasappa@....com>
---
drivers/firmware/efi/cper.c | 8 +++
drivers/firmware/efi/cper_cxl.c | 110 ++++++++++++++++++++++++++++++++
drivers/firmware/efi/cper_cxl.h | 32 ++++++++++
3 files changed, 150 insertions(+)
diff --git a/drivers/firmware/efi/cper.c b/drivers/firmware/efi/cper.c
index 1d182487fa13..5b45bf513512 100644
--- a/drivers/firmware/efi/cper.c
+++ b/drivers/firmware/efi/cper.c
@@ -623,6 +623,14 @@ cper_estatus_print_section(const char *pfx, struct acpi_hest_generic_data *gdata
cper_print_dram(newpfx, dram);
else
goto err_section_too_small;
+ } else if (guid_equal(sec_type, &CPER_SEC_CXL_MM_MODULE)) {
+ struct cper_sec_comp_event *mm_module = acpi_hest_get_payload(gdata);
+
+ printk("%ssection_type: CXL Memory Module Event\n", newpfx);
+ if (gdata->error_data_length >= sizeof(*mm_module))
+ cper_print_mm_module(newpfx, mm_module);
+ else
+ goto err_section_too_small;
} else {
const void *err = acpi_hest_get_payload(gdata);
diff --git a/drivers/firmware/efi/cper_cxl.c b/drivers/firmware/efi/cper_cxl.c
index 3fba360b7dc6..5be10ca20c7c 100644
--- a/drivers/firmware/efi/cper_cxl.c
+++ b/drivers/firmware/efi/cper_cxl.c
@@ -39,6 +39,11 @@
#define DRAM_VALID_COLUMN BIT_ULL(6)
#define DRAM_VALID_CORRECTION_MASK BIT_ULL(7)
+#define DHI_AS_LIFE_USED(as) (as & GENMASK(1, 0))
+#define DHI_AS_DEV_TEMP(as) (((as) & GENMASK(3, 2)) >> 2)
+#define DHI_AS_COR_VOL_ERR_CNT(as) (((as) & GENMASK(4, 4)) >> 4)
+#define DHI_AS_COR_PER_ERR_CNT(as) (((as) & GENMASK(5, 5)) >> 5)
+
/* CXL RAS Capability Structure, CXL v3.0 sec 8.2.4.16 */
struct cxl_ras_capability_regs {
u32 uncor_status;
@@ -119,6 +124,45 @@ static const char * const dram_mem_type_strs[] = {
"data path error",
};
+static const char * const mm_module_event_type_strs[] = {
+ "health status change",
+ "media status change",
+ "life used change",
+ "temperature change",
+ "data path error",
+ "lsa error",
+};
+
+static const char * const dhi_health_status_strs[] = {
+ "maintenance needed",
+ "performance degraded",
+ "hardware replacement needed",
+};
+
+static const char * const dhi_media_status_strs[] = {
+ "normal",
+ "not ready",
+ "write persistency lost",
+ "all data lost",
+ "write persistency loss in the event of power loss",
+ "write persistency loss in event of shutdown",
+ "write persistency loss imminent",
+ "all data loss in the event of power loss",
+ "all data loss in the event of shutdown",
+ "all data loss imminent",
+};
+
+static const char * const dhi_two_bit_status_strs[] = {
+ "normal",
+ "warning",
+ "critical",
+};
+
+static const char * const dhi_one_bit_status_strs[] = {
+ "normal",
+ "warning",
+};
+
void cper_print_prot_err(const char *pfx, const struct cper_sec_prot_err *prot_err)
{
if (prot_err->valid_bits & PROT_ERR_VALID_AGENT_TYPE)
@@ -409,3 +453,69 @@ void cper_print_dram(const char *pfx, const struct cper_sec_comp_event *event)
dram->cor_mask, sizeof(dram->cor_mask), 0);
}
}
+
+static void cper_print_mm_module_dhi(const char *pfx, const struct dev_health_info *dhi)
+{
+ pr_info("%s health status: 0x%02x\n", pfx, dhi->health_status);
+ cper_print_bits(pfx, dhi->health_status, dhi_health_status_strs,
+ ARRAY_SIZE(dhi_health_status_strs));
+
+ pr_info("%s media status: %d, %s\n", pfx, dhi->media_status,
+ dhi->media_status < ARRAY_SIZE(dhi_media_status_strs)
+ ? dhi_media_status_strs[dhi->media_status] : "unknown");
+
+ pr_info("%s current life used: %ld, %s\n", pfx,
+ DHI_AS_LIFE_USED(dhi->add_status),
+ DHI_AS_LIFE_USED(dhi->add_status) < ARRAY_SIZE(dhi_two_bit_status_strs)
+ ? dhi_two_bit_status_strs[DHI_AS_LIFE_USED(dhi->add_status)]
+ : "unknown");
+
+ pr_info("%s current device temperature: %ld, %s\n", pfx,
+ DHI_AS_DEV_TEMP(dhi->add_status),
+ DHI_AS_DEV_TEMP(dhi->add_status) < ARRAY_SIZE(dhi_two_bit_status_strs)
+ ? dhi_two_bit_status_strs[DHI_AS_DEV_TEMP(dhi->add_status)]
+ : "unknown");
+
+ pr_info("%s current corrected volatile err count: %ld, %s\n", pfx,
+ DHI_AS_COR_VOL_ERR_CNT(dhi->add_status),
+ DHI_AS_COR_VOL_ERR_CNT(dhi->add_status) < ARRAY_SIZE(dhi_one_bit_status_strs)
+ ? dhi_one_bit_status_strs[DHI_AS_COR_VOL_ERR_CNT(dhi->add_status)]
+ : "unknown");
+
+ pr_info("%s current corrected persistent err count: %ld, %s\n", pfx,
+ DHI_AS_COR_PER_ERR_CNT(dhi->add_status),
+ DHI_AS_COR_PER_ERR_CNT(dhi->add_status) < ARRAY_SIZE(dhi_one_bit_status_strs)
+ ? dhi_one_bit_status_strs[DHI_AS_COR_PER_ERR_CNT(dhi->add_status)]
+ : "unknown");
+
+ pr_info("%s life used percent: 0x%02x\n", pfx, dhi->life_used);
+ pr_info("%s device temperature degree celsius: 0x%04x\n", pfx,
+ dhi->device_temp);
+ pr_info("%s dirty shutdown count: 0x%08x\n", pfx,
+ dhi->dirty_shutdown_cnt);
+ pr_info("%s total corrected volatile error count: 0x%08x\n", pfx,
+ dhi->cor_vol_err_cnt);
+ pr_info("%s total corrected persistent error count: 0x%08x\n", pfx,
+ dhi->cor_per_err_cnt);
+}
+
+void cper_print_mm_module(const char *pfx, const struct cper_sec_comp_event *event)
+{
+ struct cper_sec_mm_module *mm_module;
+
+ cper_print_comp_event(pfx, event);
+
+ if (!(event->valid_bits & COMP_EVENT_VALID_EVENT_LOG))
+ return;
+
+ mm_module = (struct cper_sec_mm_module *)(event + 1);
+
+ cper_print_event_record(pfx, &mm_module->record);
+
+ pr_info("%s device event type: %d, %s\n", pfx, mm_module->event_type,
+ mm_module->event_type < ARRAY_SIZE(mm_module_event_type_strs)
+ ? mm_module_event_type_strs[mm_module->event_type]
+ : "unknown");
+
+ cper_print_mm_module_dhi(pfx, &mm_module->dhi);
+}
diff --git a/drivers/firmware/efi/cper_cxl.h b/drivers/firmware/efi/cper_cxl.h
index 967847b571cb..c37dd624a522 100644
--- a/drivers/firmware/efi/cper_cxl.h
+++ b/drivers/firmware/efi/cper_cxl.h
@@ -25,6 +25,11 @@
GUID_INIT(0x601DCBB3, 0x9C06, 0x4EAB, 0xB8, 0xAF, 0x4E, 0x9B, \
0xFB, 0x5C, 0x96, 0x24)
+/* CXL Memory Module Event Section */
+#define CPER_SEC_CXL_MM_MODULE \
+ GUID_INIT(0xFE927475, 0xDD59, 0x4339, 0xA5, 0x86, 0x79, 0xBA, \
+ 0xB1, 0x13, 0xB7, 0x74)
+
#pragma pack(1)
/* Compute Express Link Protocol Error Section, UEFI v2.10 sec N.2.13 */
@@ -147,10 +152,37 @@ struct cper_sec_dram {
u8 reserved[23];
};
+/*
+ * CXL Memory Module Event
+ * Device Health Information - DHI
+ * CXL rev 3.0 sec 8.2.9.8.3.1; Table 8-100
+ */
+struct dev_health_info {
+ u8 health_status;
+ u8 media_status;
+ u8 add_status;
+ u8 life_used;
+ u16 device_temp;
+ u32 dirty_shutdown_cnt;
+ u32 cor_vol_err_cnt;
+ u32 cor_per_err_cnt;
+};
+
+/* CXL Memory Module Event Record
+ * CXL rev 3.0 sec 8.2.9.2.1.3; Table 8-45
+ */
+struct cper_sec_mm_module {
+ struct common_event_record record;
+ u8 event_type;
+ struct dev_health_info dhi;
+ u8 reserved[61];
+};
+
#pragma pack()
void cper_print_prot_err(const char *pfx, const struct cper_sec_prot_err *prot_err);
void cper_print_gen_media(const char *pfx, const struct cper_sec_comp_event *event);
void cper_print_dram(const char *pfx, const struct cper_sec_comp_event *event);
+void cper_print_mm_module(const char *pfx, const struct cper_sec_comp_event *event);
#endif //__CPER_CXL_
--
2.17.1
Powered by blists - more mailing lists