I am trying to build Tensorflow on Ubuntu with the command bazel build --config=opt --cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" --config=cuda //tensorflow/tools/pip_package:build_pip_package
but I get the following error after running ./configure
:
Illegal ambiguous match on configurable attribute "deps" in //tensorflow/core/grappler/costs:utils:
@local_config_cuda//cuda:using_clang
@local_config_cuda//cuda:using_nvcc
Multiple matches are not allowed unless one is unambiguously more specialized.
ERROR: Analysis of target '//tensorflow/tools/pip_package:build_pip_package' failed; build aborted
INFO: Elapsed time: 30.960s
FAILED: Build did NOT complete successfully (91 packages loaded)
I am not sure what happened. It seems that there is a conflict between using clang and nvcc, but I don't think I've specified using nvcc anywhere during the configuration process. ./configure
should have finished without any issue.
Strangely I am not able to find any such bug report on the internet.
Could it be the case that I installed CUDA incorrectly, or is it more likely a bug with the tensorflow
configuration?
You need to use --config=cuda_clang
in your Bazel command, e.g.:
bazel build --config=opt --config=cuda_clang //tensorflow/tools/pip_package:build_pip_package
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With