[SWDEV-422195/SWDEV-440985] GPU metrics 1.6

Changes:
    - Added new GPU metrics:
      1) Violation status' (ex. PVIOL/TVIOL) accumulators
      2) XCP (Graphics Compute Partitions) statistics
      3) pcie other end recovery counter
    - CLI/API/tests changes were made accordingly

Change-Id: I589b9b1f570f25dda12d95bb501feca85da8b3bb
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
This commit is contained in:
Charis Poag committed 2024-09-27 12:04:21 -05:00
1 parent 7a557b1c50
commit 3a4abbd8c0
23 files changed
+3260 -942

No files matched your search

+73 -23
View File
@@ -727,6 +727,28 @@ struct_amdsmi_vram_usage_t._fields_ = [
]
amdsmi_vram_usage_t = struct_amdsmi_vram_usage_t
class struct_amdsmi_violation_status_t(Structure):
pass
struct_amdsmi_violation_status_t._pack_ = 1 # source:False
struct_amdsmi_violation_status_t._fields_ = [
('reference_timestamp', ctypes.c_uint64),
('violation_timestamp', ctypes.c_uint64),
('per_prochot_thrm', ctypes.c_uint64),
('per_ppt_pwr', ctypes.c_uint64),
('per_socket_thrm', ctypes.c_uint64),
('per_vr_thrm', ctypes.c_uint64),
('per_hbm_thrm', ctypes.c_uint64),
('active_prochot_thrm', ctypes.c_ubyte),
('active_ppt_pwr', ctypes.c_ubyte),
('active_socket_thrm', ctypes.c_ubyte),
('active_vr_thrm', ctypes.c_ubyte),
('active_hbm_thrm', ctypes.c_ubyte),
('PADDING_0', ctypes.c_ubyte * 3),
('reserved', ctypes.c_uint64 * 24),
]
amdsmi_violation_status_t = struct_amdsmi_violation_status_t
class struct_amdsmi_frequency_range_t(Structure):
pass
@@ -804,7 +826,9 @@ struct_pcie_metric_._fields_ = [
('pcie_replay_roll_over_count', ctypes.c_uint64),
('pcie_nak_sent_count', ctypes.c_uint64),
('pcie_nak_received_count', ctypes.c_uint64),
('reserved', ctypes.c_uint64 * 13),
('pcie_lc_perf_other_end_recovery_count', ctypes.c_uint32),
('PADDING_2', ctypes.c_ubyte * 4),
('reserved', ctypes.c_uint64 * 12),
]
struct_amdsmi_pcie_info_t._pack_ = 1 # source:False
@@ -933,7 +957,8 @@ struct_amdsmi_kfd_info_t._pack_ = 1 # source:False
struct_amdsmi_kfd_info_t._fields_ = [
('kfd_id', ctypes.c_uint64),
('node_id', ctypes.c_uint32),
('reserved', ctypes.c_uint32 * 13),
('current_partition_id', ctypes.c_uint32),
('reserved', ctypes.c_uint32 * 12),
]
amdsmi_kfd_info_t = struct_amdsmi_kfd_info_t
@@ -1102,16 +1127,6 @@ amdsmi_process_handle_t = ctypes.c_uint32
class struct_amdsmi_proc_info_t(Structure):
pass
class struct_engine_usage_(Structure):
pass
struct_engine_usage_._pack_ = 1 # source:False
struct_engine_usage_._fields_ = [
('gfx', ctypes.c_uint64),
('enc', ctypes.c_uint64),
('reserved', ctypes.c_uint32 * 12),
]
class struct_memory_usage_(Structure):
pass
@@ -1123,6 +1138,16 @@ struct_memory_usage_._fields_ = [
('reserved', ctypes.c_uint32 * 10),
]
class struct_engine_usage_(Structure):
pass
struct_engine_usage_._pack_ = 1 # source:False
struct_engine_usage_._fields_ = [
('gfx', ctypes.c_uint64),
('enc', ctypes.c_uint64),
('reserved', ctypes.c_uint32 * 12),
]
struct_amdsmi_proc_info_t._pack_ = 1 # source:False
struct_amdsmi_proc_info_t._fields_ = [
('name', ctypes.c_char * 32),
@@ -1713,6 +1738,17 @@ struct_amd_metrics_table_header_t._fields_ = [
]
amd_metrics_table_header_t = struct_amd_metrics_table_header_t
class struct_amdsmi_gpu_xcp_metrics_t(Structure):
pass
struct_amdsmi_gpu_xcp_metrics_t._pack_ = 1 # source:False
struct_amdsmi_gpu_xcp_metrics_t._fields_ = [
('gfx_busy_inst', ctypes.c_uint32 * 8),
('jpeg_busy', ctypes.c_uint16 * 32),
('vcn_busy', ctypes.c_uint16 * 4),
('gfx_busy_acc', ctypes.c_uint64 * 8),
]
class struct_amdsmi_gpu_metrics_t(Structure):
pass
@@ -1780,6 +1816,17 @@ struct_amdsmi_gpu_metrics_t._fields_ = [
('jpeg_activity', ctypes.c_uint16 * 32),
('pcie_nak_sent_count_acc', ctypes.c_uint32),
('pcie_nak_rcvd_count_acc', ctypes.c_uint32),
('accumulation_counter', ctypes.c_uint64),
('prochot_residency_acc', ctypes.c_uint64),
('ppt_residency_acc', ctypes.c_uint64),
('socket_thm_residency_acc', ctypes.c_uint64),
('vr_thm_residency_acc', ctypes.c_uint64),
('hbm_thm_residency_acc', ctypes.c_uint64),
('num_partition', ctypes.c_uint16),
('PADDING_4', ctypes.c_ubyte * 6),
('xcp_stats', struct_amdsmi_gpu_xcp_metrics_t * 8),
('pcie_lc_perf_other_end_recovery', ctypes.c_uint32),
('PADDING_5', ctypes.c_ubyte * 4),
]
amdsmi_gpu_metrics_t = struct_amdsmi_gpu_metrics_t
@@ -1852,8 +1899,7 @@ struct_amdsmi_topology_nearest_t._fields_ = [
('count', ctypes.c_uint32),
('PADDING_0', ctypes.c_ubyte * 4),
('processor_list', ctypes.POINTER(None) * 32),
('reserved', ctypes.c_uint32 * 15),
('PADDING_1', ctypes.c_ubyte * 4),
('reserved', ctypes.c_uint64 * 15),
]
amdsmi_topology_nearest_t = struct_amdsmi_topology_nearest_t
@@ -2385,6 +2431,9 @@ amdsmi_get_clock_info.argtypes = [amdsmi_processor_handle, amdsmi_clk_type_t, ct
amdsmi_get_gpu_vram_usage = _libraries['libamd_smi.so'].amdsmi_get_gpu_vram_usage
amdsmi_get_gpu_vram_usage.restype = amdsmi_status_t
amdsmi_get_gpu_vram_usage.argtypes = [amdsmi_processor_handle, ctypes.POINTER(struct_amdsmi_vram_usage_t)]
amdsmi_get_violation_status = _libraries['libamd_smi.so'].amdsmi_get_violation_status
amdsmi_get_violation_status.restype = amdsmi_status_t
amdsmi_get_violation_status.argtypes = [amdsmi_processor_handle, ctypes.POINTER(struct_amdsmi_violation_status_t)]
amdsmi_get_gpu_process_list = _libraries['libamd_smi.so'].amdsmi_get_gpu_process_list
amdsmi_get_gpu_process_list.restype = amdsmi_status_t
amdsmi_get_gpu_process_list.argtypes = [amdsmi_processor_handle, ctypes.POINTER(ctypes.c_uint32), ctypes.POINTER(struct_amdsmi_proc_info_t)]
@@ -2828,9 +2877,9 @@ __all__ = \
'amdsmi_get_soc_pstate', 'amdsmi_get_socket_handles',
'amdsmi_get_socket_info', 'amdsmi_get_temp_metric',
'amdsmi_get_threads_per_core', 'amdsmi_get_utilization_count',
'amdsmi_get_xgmi_info', 'amdsmi_get_xgmi_plpd',
'amdsmi_gpu_block_t', 'amdsmi_gpu_cache_info_t',
'amdsmi_gpu_control_counter',
'amdsmi_get_violation_status', 'amdsmi_get_xgmi_info',
'amdsmi_get_xgmi_plpd', 'amdsmi_gpu_block_t',
'amdsmi_gpu_cache_info_t', 'amdsmi_gpu_control_counter',
'amdsmi_gpu_counter_group_supported', 'amdsmi_gpu_create_counter',
'amdsmi_gpu_destroy_counter', 'amdsmi_gpu_metrics_t',
'amdsmi_gpu_read_counter', 'amdsmi_gpu_xgmi_error_status',
@@ -2883,9 +2932,9 @@ __all__ = \
'amdsmi_topo_get_p2p_status', 'amdsmi_topology_nearest_t',
'amdsmi_utilization_counter_t',
'amdsmi_utilization_counter_type_t', 'amdsmi_vbios_info_t',
'amdsmi_version_t', 'amdsmi_voltage_metric_t',
'amdsmi_voltage_type_t', 'amdsmi_vram_info_t',
'amdsmi_vram_type_t', 'amdsmi_vram_usage_t',
'amdsmi_version_t', 'amdsmi_violation_status_t',
'amdsmi_voltage_metric_t', 'amdsmi_voltage_type_t',
'amdsmi_vram_info_t', 'amdsmi_vram_type_t', 'amdsmi_vram_usage_t',
'amdsmi_vram_vendor_type_t', 'amdsmi_xgmi_info_t',
'amdsmi_xgmi_status_t', 'processor_type_t', 'size_t',
'struct__links', 'struct_amd_metrics_table_header_t',
@@ -2901,6 +2950,7 @@ __all__ = \
'struct_amdsmi_freq_volt_region_t', 'struct_amdsmi_frequencies_t',
'struct_amdsmi_frequency_range_t', 'struct_amdsmi_fw_info_t',
'struct_amdsmi_gpu_cache_info_t', 'struct_amdsmi_gpu_metrics_t',
'struct_amdsmi_gpu_xcp_metrics_t',
'struct_amdsmi_hsmp_metrics_table_t', 'struct_amdsmi_kfd_info_t',
'struct_amdsmi_link_id_bw_type_t', 'struct_amdsmi_link_metrics_t',
'struct_amdsmi_name_value_t', 'struct_amdsmi_od_vddc_point_t',
@@ -2918,9 +2968,9 @@ __all__ = \
'struct_amdsmi_topology_nearest_t',
'struct_amdsmi_utilization_counter_t',
'struct_amdsmi_vbios_info_t', 'struct_amdsmi_version_t',
'struct_amdsmi_vram_info_t', 'struct_amdsmi_vram_usage_t',
'struct_amdsmi_xgmi_info_t', 'struct_cache_',
'struct_engine_usage_', 'struct_fw_info_list_',
'struct_amdsmi_violation_status_t', 'struct_amdsmi_vram_info_t',
'struct_amdsmi_vram_usage_t', 'struct_amdsmi_xgmi_info_t',
'struct_cache_', 'struct_engine_usage_', 'struct_fw_info_list_',
'struct_memory_usage_', 'struct_nps_flags_',
'struct_pcie_metric_', 'struct_pcie_static_',
'struct_amdsmi_bdf_t','uint32_t', 'uint64_t', 'uint8_t',