[SWDEV-230863] add two new interfaces for background health check (#4)

1. Get the bad pages threshold of a processor.
2. Verify the checksum of RAS EEPROM

Signed-off-by: Meng Li <li.meng@amd.com>
This commit is contained in:
Meng, Li (Jassmine)
2025-01-08 07:26:55 +08:00
committed by GitHub
parent d32f2a109a
commit dc400d916e
4 changed files with 124 additions and 0 deletions
+2
View File
@@ -45,6 +45,8 @@ amdsmi_status_t smi_amdgpu_get_power_cap(amd::smi::AMDSmiGPUDevice* device, int
amdsmi_status_t smi_amdgpu_get_ranges(amd::smi::AMDSmiGPUDevice* device, amdsmi_clk_type_t domain, int *max_freq, int *min_freq, int *num_dpm, int *sleep_state_freq);
amdsmi_status_t smi_amdgpu_get_enabled_blocks(amd::smi::AMDSmiGPUDevice* device, uint64_t *enabled_blocks);
amdsmi_status_t smi_amdgpu_get_bad_page_info(amd::smi::AMDSmiGPUDevice* device, uint32_t *num_pages, amdsmi_retired_page_record_t *info);
amdsmi_status_t smi_amdgpu_get_bad_page_threshold(amd::smi::AMDSmiGPUDevice* device, uint32_t *threshold);
amdsmi_status_t smi_amdgpu_validate_ras_eeprom(amd::smi::AMDSmiGPUDevice* device);
amdsmi_status_t smi_amdgpu_get_ecc_error_count(amd::smi::AMDSmiGPUDevice* device, amdsmi_error_count_t *err_cnt);
amdsmi_status_t smi_amdgpu_get_driver_version(amd::smi::AMDSmiGPUDevice* device, int *length, char *version);
amdsmi_status_t smi_amdgpu_get_pcie_speed_from_pcie_type(uint16_t pcie_type, uint32_t *pcie_speed);