On this page |
Overview ¶
Some Houdini features and nodes require CUDA and cuDNN, which are third-party tools. CUDA provides access to a development environment for general computing on GPUs, while cuDNN accelerates Deep Learning models.
Use the following workflows to install CUDA and cuDNN on Windows or Linux. See Troubleshooting if you encounter issues while installing the tools.
Note
Installing CUDA is a prerequisite for installing cuDNN.
Installing the CUDA Toolkit ¶
-
Determine the latest version of CUDA that Houdini supports (see Houdini third-party libraries and tools).
-
Use one of the following installation guides to install the CUDA Toolkit on your system.
Note
Make sure you use the correct CUDA version in your installation. For example, use
cuda-toolkit-12-2
instead ofcuda-toolkit-12-6
if that’s the supported CUDA version. -
Restart Houdini. You should now be able to use features that require CUDA.
Installing cuDNN ¶
-
Determine the latest version of cuDNN that Houdini supports (see Houdini third-party libraries and tools).
-
Use one of the following installation guides to install cuDNN on your system.
Note
Make sure you use the correct cuDNN and CUDA versions in your installation. For example, use
libcudnn8
instead oflibcudnn9
if that’s the supported cuDNN version. -
Restart Houdini. You should now be able to use features that require cuDNN.
Troubleshooting ¶
Windows
-
You may receive a warning that says
Suspiciously long error messages truncated
. This warning is usually caused by CUDA running out of memory. Restarting Houdini may resolve this issue. -
You may receive a warning that says
Failed to load any of the following: "cublas64_12.dll", "cublasLt64_12.dll", "cufft64_11.dll", "curand64_10.dll"
. This warning is usually caused by the CUDA installation. Make sure you install the correct version of CUDA to prevent this issue. Runnvcc --version
in the command line or power shell to verify your installation. -
You may receive a warning that says
Failed to load cudnn64_8.dll
. This warning is usually caused by issues with the cuDNN installation. Make sure you install cuDNN in the correct location and that the PATH variable points to a folder containing the.dll
file. -
You may receive a warning that says
Failed to load cudnn_cnn_infer64_8.dll
. This warning can be caused by an issue with the cuDNN installation. Make sure you install cuDNN in the correct location.
Linux
-
You may receive a warning that says
Suspiciously long error messages truncated
. This warning is usually caused by CUDA running out of memory. Restarting Houdini may resolve this issue. -
You may receive a warning that says
Failed to load any of the following: "libcublas.so.12", "libcublasLt.so.12", "libcufft.so.11", "libcurand.so.10"
. This warning is usually caused by the CUDA installation. Make sure you install the correct version of CUDA to prevent this issue. Runnvcc --version
in the command line to verify your installation. -
You may receive a warning that says
Failed to load libcudnn.so.8
. This warning is usually caused by issues with the cuDNN installation. Make sure you install cuDNN in the correct location. -
You may receive a warning that says
Failed to load libcudnn_cnn_infer.so.8
. This warning can be caused by either an issue with the cuDNN or zlib installation. Make sure you install cuDNN and zlib in the correct locations.