diff options
| author | Tor Aamodt <[email protected]> | 2019-05-30 07:49:58 -0700 |
|---|---|---|
| committer | GitHub <[email protected]> | 2019-05-30 07:49:58 -0700 |
| commit | 6337105df018f25ad1b58afa040d8b1006f2a0ce (patch) | |
| tree | fd45030f1adc6b60ca77c7ecd5af89aebb7e1c03 | |
| parent | 5837143ef70e20c362083b36c55e4c473f59f358 (diff) | |
| parent | 6d99e17334235b9e94abb82b0aee66c78f737748 (diff) | |
Merge pull request #129 from PSuchita/dev
Updating README
| -rw-r--r-- | README | 26 |
1 files changed, 25 insertions, 1 deletions
@@ -177,7 +177,7 @@ GPGPU-Sim dependencies: * flex * zlib * CUDA Toolkit - + GPGPU-Sim documentation dependencies: * doxygen * graphvi @@ -208,6 +208,9 @@ python-matplotlib" CUDA SDK dependencies: "sudo apt-get install libxi-dev libxmu-dev libglut3-dev" +If you are running applications which use NVIDIA libraries such as cuDNN and +cuBLAS, install them too. + Finally, ensure CUDA_INSTALL_PATH is set to the location where you installed the CUDA Toolkit (e.g., /usr/local/cuda) and that $CUDA_INSTALL_PATH/bin is in your PATH. You probably want to modify your .bashrc file to incude the @@ -216,6 +219,10 @@ following (this assumes the CUDA Toolkit was installed in /usr/local/cuda): export CUDA_INSTALL_PATH=/usr/local/cuda export PATH=$CUDA_INSTALL_PATH/bin +If running applications which use cuDNN or cuBLAS: + export CUDNN_PATH=<Path To cuDNN Directory> + export LD_LIBRARY_PATH=$CUDA_INSTALL_PATH/lib64:$CUDA_INSTALL_PATH/lib:$CUDNN_PATH/lib64 + Step 2: Build ============= @@ -261,6 +268,23 @@ ldd <your_application_name> You should see that your application is using libcudart.so file in GPGPUSim directory. +If running applications which use cuDNN or cuBLAS: + + * Modify the Makefile or the compilation command of the application to change + all the dynamic links to static ones, for example: + * -L$(CUDA_PATH)/lib64 -lcublas to + -L$(CUDA_PATH)/lib64 -lcublas_static + + * -L$(CUDNN_PATH)/lib64 -lcudnn to + -L$(CUDNN_PATH)/lib64 -lcudnn_static + + * Modify the Makefile or the compilation command such that the following + flags are used by the nvcc compiler: + -gencode arch=compute_61,code=compute_61 + + (the number 61 refers to the SM version. You would need to set it based + on the GPGPU-Sim config "-gpgpu ptx force max capability" you use) + Copy the contents of configs/QuadroFX5800/ or configs/GTX480/ to your application's working directory. These files configure the microarchitecture models to resemble the respective GPGPU architectures. |