Useful or not, from you.
tensorflow tensorflow-nightly-gpu looking for cusolver64_10.dll on a cuDNN 11.1 installation

<em>Please make sure that this is a build/installation issue. As per our GitHub Policy, we only address code/doc bugs, performance issues, feature requests and build/installation issues on GitHub. tag:build_template</em>

System information

  • OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Windows 10
  • Mobile device (e.g. iPhone 8, Pixel 2, Samsung Galaxy) if the issue happens on mobile device:
  • TensorFlow installed from (source or binary): binary
  • TensorFlow version: tf-nightly-gpu-2.4.0.dev20201023
  • Python version: 3.7.9
  • Installed using virtualenv? pip? conda?: pip
  • Bazel version (if compiling from source):
  • GCC/Compiler version (if compiling from source):
  • CUDA/cuDNN version: 11.1
  • GPU model and memory: GeForceMX250

Describe the problem After installing the new CUDA and cuDNN, I can now get the installation to go through. However, the first command tensorflow.test.is_gpu_available() fails because one dll is not found:

2020-10-24 11:59:36.156976: W tensorflow/stream_executor/platform/default/] Could not load dynamic library 'cusolver64_10.dll'; dlerror: cusolver64_10.dll not found

In the "NVIDIA GPU Computing Toolkit"\CUDA\v11.1\bin directory I see the following dll:


However, tensorflow nightly seems to be looking for cusolver64_10.dll.

Provide the exact sequence of commands / steps that you executed before running into the problem python import tensorflow tensorflow.test.is_gpu_available()

Any other info / logs Include any logs or source code that would be helpful to diagnose the problem. If including tracebacks, please include the full traceback. Large logs and files should be attached.

That's a useful answer
Without any help

Please use CUDA 11.0 when using TF 2.4 (and nightly). We've built and tested against CUDA 11.0, not 11.1.

Installation of CUDA 10.2 solved the problem.