[SWDEV-230863] Improve the functionality of RdcSmiHealth module.

Memory check:get the threshold of retired page number
EEPROM check:read and verify the checksum
Power/Thermal check: power/thermal throttle status counter

Signed-off-by: Meng Li <li.meng@amd.com>
Change-Id: Id2c751416eb5bf007e6e1da8dc05966a6ba1324e
Этот коммит содержится в:
limeng12
2025-01-09 13:52:56 +08:00
коммит произвёл Meng, Li (Jassmine)
родитель 83f36f1673
Коммит 016a1d9d39
11 изменённых файлов: 307 добавлений и 72 удалений
+1 -1
Просмотреть файл
@@ -181,7 +181,7 @@ rdc_status_t RdcSmiLib::rdc_telemetry_fields_query(uint32_t field_ids[MAX_NUM_FI
RDC_EVNT_XGMI_4_THRPUT, RDC_EVNT_XGMI_5_THRPUT, RDC_FI_OAM_ID,
RDC_FI_GPU_MM_ENC_UTIL, RDC_FI_GPU_MM_DEC_UTIL, RDC_FI_GPU_MEMORY_ACTIVITY,
RDC_HEALTH_XGMI_ERROR, RDC_HEALTH_PCIE_REPLAY_COUNT, RDC_HEALTH_RETIRED_PAGE_NUM,
RDC_HEALTH_PENDING_PAGE_NUM, RDC_HEALTH_RETIRED_PAGE_LIMIT, RDC_HEALTH_UNCORRECTABLE_PAGE_LIMIT,
RDC_HEALTH_PENDING_PAGE_NUM, RDC_HEALTH_RETIRED_PAGE_LIMIT, RDC_HEALTH_EEPROM_CONFIG_VALID,
RDC_HEALTH_POWER_THROTTLE_TIME, RDC_HEALTH_THERMAL_THROTTLE_TIME,
RDC_FI_GPU_MEMORY_MAX_BANDWIDTH, RDC_FI_GPU_MEMORY_CUR_BANDWIDTH,
};