diff options
author | 2016-05-17 13:42:26 -0800 | |
---|---|---|
committer | 2016-05-17 14:52:08 -0700 | |
commit | 4d55cb5c55fcc85009171b6a4657cbd966fd85e5 (patch) | |
tree | 303d69d5be3054475262d49ceefc9c116ec5da40 /tensorflow/python/util/port.i | |
parent | 2709b2303b39b848b9c6a86cd89811820d390580 (diff) |
Enable fp16 support for MatMul via cuBLAS, gated on compilation with
CUDA 7.5 or higher. (If not, the GPU tests will also not be run,
so that the tests as a whole will keep passing on CUDA 7.0.)
Change: 122566230
Diffstat (limited to 'tensorflow/python/util/port.i')
-rw-r--r-- | tensorflow/python/util/port.i | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/tensorflow/python/util/port.i b/tensorflow/python/util/port.i index 568658cd7f..c3c833faef 100644 --- a/tensorflow/python/util/port.i +++ b/tensorflow/python/util/port.i @@ -22,5 +22,6 @@ limitations under the License. %ignoreall %unignore tensorflow; %unignore tensorflow::IsGoogleCudaEnabled; +%unignore tensorflow::CudaSupportsHalfMatMulAndConv; %include "tensorflow/core/util/port.h" %unignoreall |