Fix missing dash for cxxopt build option (#9736)

This commit is contained in:
Mortada Mehyar 2017-05-06 23:22:17 -07:00 committed by Vijay Vasudevan
parent 6ee900438c
commit d156525025

View File

@ -42,7 +42,7 @@ To incorporate your custom op you'll need to:
Python @{tf.test.compute_gradient_error$gradient checker}. Python @{tf.test.compute_gradient_error$gradient checker}.
See See
[`relu_op_test.py`](https://www.tensorflow.org/code/tensorflow/python/kernel_tests/relu_op_test.py) as [`relu_op_test.py`](https://www.tensorflow.org/code/tensorflow/python/kernel_tests/relu_op_test.py) as
an example that does tests the forward functions of Relu-like operators and an example that tests the forward functions of Relu-like operators and
their gradients. their gradients.
PREREQUISITES: PREREQUISITES:
@ -188,7 +188,7 @@ building the `.so` file.
> the older ABI. If you compile your op library with `gcc>=5`, add > the older ABI. If you compile your op library with `gcc>=5`, add
> `-D_GLIBCXX_USE_CXX11_ABI=0` to the command line to make the library > `-D_GLIBCXX_USE_CXX11_ABI=0` to the command line to make the library
> compatible with the older abi. > compatible with the older abi.
> Furthermore if you are using TensorFlow package created from source remember to add `-cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0"` > Furthermore if you are using TensorFlow package created from source remember to add `--cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0"`
> as bazel command to compile the Python package. > as bazel command to compile the Python package.
### Compile the op using bazel (TensorFlow source installation) ### Compile the op using bazel (TensorFlow source installation)