search

apache / mxnet

Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more
https://mxnet.apache.org
Apache License 2.0
20.77k stars 6.79k forks source link

nvcc fatal unsupported gpu architecture 'compute_86' #20368

Closed Adnios closed 3 years ago

Adnios commented 3 years ago

Description

When I install mxnet from source, it gives me nvcc fatal unsupported gpu architecture 'compute_86'. My cuda version is 11.0. I can alleviate the issue by change Auto to ;8.0. https://github.com/apache/incubator-mxnet/blob/b54b7f36c944b933f2aa94f224821548da5fcaf6/CMakeLists.txt#L43

I think the current mxnet installation may not handle cuda 11.0 well.

To Reproduce

rm -rf build
mkdir -p build && cd build
cmake -GNinja \
    -DUSE_CUDA=ON \
    -DUSE_MKL_IF_AVAILABLE=ON \
    -DCMAKE_CUDA_COMPILER_LAUNCHER=ccache \
    -DCMAKE_C_COMPILER_LAUNCHER=ccache \
    -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
    -DCMAKE_BUILD_TYPE=Release \
..
ninja
github-actions[bot] commented 3 years ago

Welcome to Apache MXNet (incubating)! We are on a mission to democratize AI, and we are glad that you are contributing to it by opening this issue. Please make sure to include all the relevant context, and one of the @apache/mxnet-committers will be here shortly. If you are interested in contributing to our project, let us know! Also, be sure to check out our guide on contributing to MXNet and our development guides wiki.

leezu commented 3 years ago

This is caused by outdated copy of https://gitlab.kitware.com/cmake/cmake/blob/master/Modules/FindCUDA/select_compute_arch.cmake Should be fixed by https://github.com/apache/incubator-mxnet/pull/20369

Adnios commented 3 years ago

Thanks @leezu . I think this issus is fixed by the https://github.com/apache/incubator-mxnet/pull/20369. Closed. image