Inference [4.1.5] Inference nvidia-smi command exits but failed to execute

Hello,

I ran the following command: nohup make generate_engines RUN_ARGS="--benchmarks=stable-diffusion-xl --scenarios=offline" 2>&1 | tee -a /Inference_Store/mlperf_4.1.5_inference_data/mlperf-logs/generate_engines_offline

And ran into the following errors.
[2025-02-06 15:29:21,626 systems.py:174 WARNING] nvidia-smi command exists but failed to execute - Ignoring NUMA detection.
[2025-02-06 15:29:21,870 main.py:227 INFO] Detected system did not match any known systems. Exiting. SystemConfiguration(host_cpu_conf=CPUConfiguration(layout={CPU(name='Cortex-A78AE', architecture=<CPUArchitecture.aarch64: AliasedName(name='aarch64', aliases=(), patterns=())>, core_count=4, threads_per_core=1): 1}), host_mem_conf=MemoryConfiguration(host_memory_capacity=Memory(quantity=64.348844, byte_suffix=<ByteSuffix.GB: (1000, 3)>, _num_bytes=64348844000), comparison_tolerance=0.05), accelerator_conf=AcceleratorConfiguration(layout=defaultdict(<class 'int'>, {GPU(name='CTI Forge + Orin AGX', accelerator_type=<AcceleratorType.Integrated: AliasedName(name='Integrated', aliases=(), patterns=())>, vram=None, max_power_limit=None, pci_id=None, compute_sm=87): 1})), numa_conf=None, system_id=None)
(mlperf) nvidia@mlperf-inference-nvidia-aarch64-1018:/closed/NVIDIA$ nvidia-smi
Thu Feb  6 15:34:15 2025

Any advice on how to resolve this issue? How can I ensure my system is detected and how can I check what my known systems are.

NVIDIA-SMI version: 540.4.0
Cuda version: 12.6
Driver version 540.4.0

Start with systems.py:174, where the nvidia-smi warning is emitted, and main.py:227, where unmatched system configurations exit. Reproduce the make generate_engines command and inspect how known systems are represented and selected for the reported CPU, memory, and accelerator configuration. Done means the reported system is either detected correctly or the unsupported case is clearly handled.