[문제해결] NUMA node read from SysFS had negative value -1
reference
https://lapina.tistory.com/137
successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero.
>>> import tensorflow as tf
>>> from tensorflow.python.client import device_lib
>>> tf.test.is_gpu_available()# 사용 가능한 GPU가 있는지 확인: True
WARNING:tensorflow:From <stdin>:1: is_gpu_available (from tensorflow.python.framework.test_util) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.config.list_physical_devices('GPU')` instead.
2021-08-13 09:59:51.226432: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2021-08-13 09:59:51.277631: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-08-13 09:59:51.283808: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-08-13 09:59:51.284694: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-08-13 09:59:51.543833: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-08-13 09:59:51.544531: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-08-13 09:59:51.545183: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-08-13 09:59:51.545830: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1505] Created device /device:GPU:0 with 21984 MB memory: -> device: 0, name: NVIDIA GeForce RTX 3090, pci bus id: 0000:01:00.0, compute capability: 8.6
True
tf-nighly GPU를 활용하기 위해 환경설정 하는 중 위와 같은 메시지가 뜨는 현상이 발생했다.
알아보니,
NUMA node 정보가 올바르지 않지만, 너 적어도 한 개의 NUMA node 있다.
그러니 일단 되게끔 해 보겠다.
라는 메시지로 요약된다.