CODEWITHSUNDEEP
pythonlinuxubuntutensorflow
Daniel Paczuski Bak
Daniel Paczuski Bak

Reputation: 4088

Bazel build tensorflow: The specified --crosstool_top '//third_party/gpus/crosstool:crosstool' is not a valid cc_toolchain_suite rule

I am following the steps outlined here: http://www.nvidia.com/object/gpu-accelerated-applications-tensorflow-installation.html

I have a GTX 860M, and Ubuntu 16.04.

I followed the instructions precisely, except for Step 5. Where I pointed to my actual python binary location (/usr/bin/python2.7).

Here's the error I'm getting:

daniel@beepboop:~/tensorflow$ bazel build -c opt --config=cuda //tensorflow/tools/pip_package:build_pip_package
WARNING: /home/daniel/tensorflow/tensorflow/tensorflow.bzl:477:26: Variables HOST_CFG and DATA_CFG are deprecated in favor of strings "host" and "data" correspondingly.
WARNING: /home/daniel/tensorflow/tensorflow/tensorflow.bzl:490:24: Variables HOST_CFG and DATA_CFG are deprecated in favor of strings "host" and "data" correspondingly.
ERROR: The specified --crosstool_top '//third_party/gpus/crosstool:crosstool' is not a valid cc_toolchain_suite rule.

What am I missing here?

Upvotes: 2

Views: 1078

Answers (2)

Xinyu Wu
Xinyu Wu

Reputation: 11

Drop this line "$ git reset --hard 70de76e", and select "N" for all questions except the one for GPU support as following:

$ ./configure
Do you wish to build TensorFlow with Google Cloud Platform support? [y/N] n
Do you wish to build TensorFlow with Hadoop File System support? [y/N] n
Do you wish to build TensorFlow with OpenCL support? [y/N] n
Do you wish to build TensorFlow with GPU support? [y/N] y

Upvotes: 1

Daniel Paczuski Bak
Daniel Paczuski Bak

Reputation: 4088

Not doing the git reset in step 5 caused everything to work correctly.

Upvotes: 1

Related Questions