fix: [rocm/amd_smi_lib] amdsmi_get_gpu_activity gfx/memory activity does not update

Checks and forces rereading gpu metrics unconditionally

Code changes related to the following:
  * Device::dev_log_gpu_metrics()
  * amdsmi_get_gpu_metrics_header_info()
    Removed unintentionally during work on 'header cleanup Remove non-unified headers'
  * Examples
  * Unit tests

Change-Id: I83710e173c0f7102d0b7f865c18474c979a95cd8
Signed-off-by: Oliveira, Daniel <daniel.oliveira@amd.com>
This commit is contained in:
Oliveira, Daniel
2024-02-09 15:35:24 -06:00
committed by Maisam Arif
parent f831cf49f7
commit 78074d7d77
6 changed files with 334 additions and 113 deletions
+17 -13
View File
@@ -746,19 +746,6 @@ amdsmi_card_form_factor_t = ctypes.c_uint32 # enum
class struct_amdsmi_pcie_info_t(Structure):
pass
class struct_pcie_static_(Structure):
pass
struct_pcie_static_._pack_ = 1 # source:False
struct_pcie_static_._fields_ = [
('max_pcie_lanes', ctypes.c_uint16),
('PADDING_0', ctypes.c_ubyte * 2),
('max_pcie_speed', ctypes.c_uint32),
('pcie_interface_version', ctypes.c_uint32),
('slot_type', amdsmi_card_form_factor_t),
('reserved', ctypes.c_uint64 * 10),
]
class struct_pcie_metric_(Structure):
pass
@@ -777,6 +764,19 @@ struct_pcie_metric_._fields_ = [
('reserved', ctypes.c_uint64 * 13),
]
class struct_pcie_static_(Structure):
pass
struct_pcie_static_._pack_ = 1 # source:False
struct_pcie_static_._fields_ = [
('max_pcie_lanes', ctypes.c_uint16),
('PADDING_0', ctypes.c_ubyte * 2),
('max_pcie_speed', ctypes.c_uint32),
('pcie_interface_version', ctypes.c_uint32),
('slot_type', amdsmi_card_form_factor_t),
('reserved', ctypes.c_uint64 * 10),
]
struct_amdsmi_pcie_info_t._pack_ = 1 # source:False
struct_amdsmi_pcie_info_t._fields_ = [
('pcie_static', struct_pcie_static_),
@@ -1993,6 +1993,9 @@ amdsmi_reset_gpu.argtypes = [amdsmi_processor_handle]
amdsmi_get_gpu_od_volt_info = _libraries['libamd_smi.so'].amdsmi_get_gpu_od_volt_info
amdsmi_get_gpu_od_volt_info.restype = amdsmi_status_t
amdsmi_get_gpu_od_volt_info.argtypes = [amdsmi_processor_handle, ctypes.POINTER(struct_amdsmi_od_volt_freq_data_t)]
amdsmi_get_gpu_metrics_header_info = _libraries['libamd_smi.so'].amdsmi_get_gpu_metrics_header_info
amdsmi_get_gpu_metrics_header_info.restype = amdsmi_status_t
amdsmi_get_gpu_metrics_header_info.argtypes = [amdsmi_processor_handle, ctypes.POINTER(struct_amd_metrics_table_header_t)]
amdsmi_get_gpu_metrics_info = _libraries['libamd_smi.so'].amdsmi_get_gpu_metrics_info
amdsmi_get_gpu_metrics_info.restype = amdsmi_status_t
amdsmi_get_gpu_metrics_info.argtypes = [amdsmi_processor_handle, ctypes.POINTER(struct_amdsmi_gpu_metrics_t)]
@@ -2527,6 +2530,7 @@ __all__ = \
'amdsmi_get_gpu_id', 'amdsmi_get_gpu_memory_partition',
'amdsmi_get_gpu_memory_reserved_pages',
'amdsmi_get_gpu_memory_total', 'amdsmi_get_gpu_memory_usage',
'amdsmi_get_gpu_metrics_header_info',
'amdsmi_get_gpu_metrics_info',
'amdsmi_get_gpu_od_volt_curve_regions',
'amdsmi_get_gpu_od_volt_info', 'amdsmi_get_gpu_overdrive_level',