1

I'm attempting to use Tensorflow with a Rtx 3090 GPU, however I've been experiencing a variety of issues for several days. I tried the remedies suggested here and in other places, but they didn't work. Either a kernel error occurs, or the program proceeds with the CPU without seeing the GPU. Could you please assist me?

2021 13:21:07.654550: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cudart64_110.dll
2021 13:21:09.144192: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance‑critical operations: AVX AVX2
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2021 13:21:09.149726: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library nvcuda.dll
2021 13:21:09.172491: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1733] Found device 0 with properties: 
pciBusID: 0000:08:00.0 name: NVIDIA GeForce RTX 3090 computeCapability: 8.6
coreClock: 1.74GHz coreCount: 82 deviceMemorySize: 24.00GiB deviceMemoryBandwidth: 871.81GiB/s
2021 13:21:09.173145: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cudart64_110.dll
2021 13:21:09.201143: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cublas64_11.dll
2021 13:21:09.201496: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cublasLt64_11.dll
2021 13:21:09.218490: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cufft64_10.dll
2021 13:21:09.222724: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library curand64_10.dll
2021 13:21:09.253841: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cusolver64_11.dll
2021 13:21:09.272022: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cusparse64_11.dll
2021 13:21:09.272867: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cudnn64_8.dll
2021 13:21:09.273229: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1871] Adding visible gpu devices: 0
2021 13:21:09.715332: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1258] Device interconnect StreamExecutor with strength 1 edge matrix:
2021 13:21:09.715688: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1264] 0 
2021 13:21:09.715891: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1277] 0: N 
2021 13:21:09.716223: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1418] Created TensorFlow device (/device:GPU:0 with 18786 MB memory) ‑> physical GPU (device: 0, name: NVIDIA GeForce RTX 3090, pci bus id: 0000:08:00.0, compute capability: 8.6)
2021 13:21:10.046619: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1733] Found device 0 with properties: 
pciBusID: 0000:08:00.0 name: NVIDIA GeForce RTX 3090 computeCapability: 8.6
coreClock: 1.74GHz coreCount: 82 deviceMemorySize: 24.00GiB deviceMemoryBandwidth: 871.81GiB/s
2021 13:21:10.047281: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1871] Adding visible gpu devices: 0
2021 13:21:10.047754: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1733] Found device 0 with properties: 
pciBusID: 0000:08:00.0 name: NVIDIA GeForce RTX 3090 computeCapability: 8.6
coreClock: 1.74GHz coreCount: 82 deviceMemorySize: 24.00GiB deviceMemoryBandwidth: 871.81GiB/s
2021 13:21:10.048414: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1871] Adding visible gpu devices: 0
2021 13:21:10.048707: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1258] Device interconnect StreamExecutor with strength 1 edge matrix:
2021 13:21:10.049027: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1264] 0 
2021 13:21:10.049227: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1277] 0: N 
2021 13:21:10.049491: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1418] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 18786 MB memory) ‑> physical GPU (device: 0, name: NVIDIA GeForce RTX 3090, pci bus id: 0000:08:00.0, compute capability: 8.6)
2021 13:21:10.928282: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:176] None of the MLIR Optimization Passes are enabled (registered 2)
2021 13:21:25.315947: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cudnn64_8.dll
Tsyvarev
  • 60,011
  • 17
  • 110
  • 153
Selçuk Aktaş
  • 11
  • 2
  • You may want to look at this answer. Did you do all these steps? https://stackoverflow.com/a/51307381/3961841 – Ladislav Ondris Dec 23 '21 at 10:40
  • Yes, I followed all of the directions for that as well as other concerns. However, I was unable to run it on Tensorflow with a GPU since it continually consumes the CPU. I was stumped as to what the issue was. – Selçuk Aktaş Dec 23 '21 at 13:45

1 Answers1

2

These are just informational messages as they are prefixed with I, if it is the error message they would be prefixed with E or W for warnings are as shown below:

2020-12-30 21:30:27.549172: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cupti64_101.dll

2020-12-30 21:30:27.599977: W tensorflow/core/framework/allocator.cc:101] Allocation of 37171200 exceeds 10% of system memory.

2021-12-30 21:30:27.704083: E tensorflow/core/profiler/internal/gpu/cupti_tracer.cc:1307] function cupti_interface_->Subscribe( &subscriber_, (CUpti_CallbackFunc)ApiCallback, this)failed with error CUPTI_ERROR_INSUFFICIENT_PRIVILEGES

You can surpass these warnings using below code:

import os
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'

You can also check executing this code:

import tensorflow as tf
print("Num GPUs Available: ", len(tf.config.list_physical_devices('GPU')))