diff options
author | Tatiana Shpeisman <shpeisman@google.com> | 2018-03-05 14:40:46 -0800 |
---|---|---|
committer | TensorFlower Gardener <gardener@tensorflow.org> | 2018-03-05 14:44:58 -0800 |
commit | 06c7a190ac122512edf7229041f34391d8993da0 (patch) | |
tree | 7e5c5597efffd89efe266804de9f1435102df9f1 /tensorflow/core/util/port.h | |
parent | 0e9289489f9dac926b7de5eae47417daad6d626f (diff) |
Adds test_util.IsMklEnabled() that returns true if TensorFlow has been built with MKL support. Fixes the failure of tensorflow/python/tools:print_selective_registration_header_test by using 'Mkl' prefix for MatMul op name when MKL is enabled.
PiperOrigin-RevId: 187925038
Diffstat (limited to 'tensorflow/core/util/port.h')
-rw-r--r-- | tensorflow/core/util/port.h | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/tensorflow/core/util/port.h b/tensorflow/core/util/port.h index ed65341711..981def9d22 100644 --- a/tensorflow/core/util/port.h +++ b/tensorflow/core/util/port.h @@ -25,6 +25,9 @@ bool IsGoogleCudaEnabled(); // half-precision matrix multiplications and convolution operations. bool CudaSupportsHalfMatMulAndConv(); +// Returns true if INTEL_MKL is defined +bool IsMklEnabled(); + } // end namespace tensorflow #endif // TENSORFLOW_UTIL_PORT_H_ |