zachgk commented on a change in pull request #13364: [MXNET-1225] Always use config.mk in make install instructions URL: https://github.com/apache/incubator-mxnet/pull/13364#discussion_r238521670
########## File path: docs/install/build_from_source.md ########## @@ -214,57 +214,79 @@ It is recommended to set environment variable NCCL_LAUNCH_MODE to PARALLEL when For example, you can specify using all cores on Linux as follows: ```bash -cmake -j$(nproc) +cmake -j $(nproc) . ``` #### Recommended for Systems with NVIDIA GPUs and Intel CPUs * Build MXNet with `cmake` and install with MKL DNN, GPU, and OpenCV support: ```bash -cmake -j USE_CUDA=1 USE_CUDA_PATH=/usr/local/cuda USE_CUDNN=1 USE_MKLDNN=1 +echo "USE_OPENCV = 1" >> ./config.mk +echo "USE_CUDA = 1" >> ./config.mk +echo "USE_CUDA_PATH = /usr/local/cuda" >> ./config.mk +echo "USE_CUDNN = 1" >> ./config.mk +echo "USE_MKLDNN = 1" >> ./config.mk +cmake -j $(nproc) USE_CUDA=1 USE_CUDA_PATH=/usr/local/cuda USE_CUDNN=1 USE_MKLDNN=1 . Review comment: @larroy, I was testing it on an EC2 and it seems to work because -j was added in cmake 3.12 (https://cmake.org/cmake/help/v3.12/release/3.12.html#cmake-3-12-release-notes). But, it may be better to do something compatible with older versions of cmake. What is the correct way that we should be invoking cmake? ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
