diff options
| author |  Benoit Steiner <benoit.steiner.goog@gmail.com> | 2016-03-03 10:37:25 -0800 |
|---|---|---|
| committer | Benoit Steiner <benoit.steiner.goog@gmail.com> | 2016-03-03 10:37:25 -0800 |
| commit | dac58d7c3599a1c5c7631a734ca95a60ddb549ef (patch) | |
| tree | bc915db94ab96b20dd30501af3b2954e77278e21 /unsupported/test/CMakeLists.txt | |
| parent | 1032441c6fea0a0d98b394abe8ffdb228256f47b (diff) | |
Added a test to validate the conversion of half floats into floats on Kepler GPUs.
Restricted the testing of the random number generation code to GPU architecture greater than or equal to 3.5.
Diffstat (limited to 'unsupported/test/CMakeLists.txt')
| -rw-r--r-- | unsupported/test/CMakeLists.txt | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/unsupported/test/CMakeLists.txt b/unsupported/test/CMakeLists.txt index 3be43f47f..bc9248b9e 100644 --- a/unsupported/test/CMakeLists.txt +++ b/unsupported/test/CMakeLists.txt @@ -170,10 +170,15 @@ if(CUDA_FOUND) ei_add_test(cxx11_tensor_cuda) ei_add_test(cxx11_tensor_contract_cuda) ei_add_test(cxx11_tensor_reduction_cuda) - ei_add_test(cxx11_tensor_random_cuda) ei_add_test(cxx11_tensor_argmax_cuda) + ei_add_test(cxx11_tensor_cast_float16_cuda) - # Half floats are only supported starting with arch 5.3 + # The random number generation code requires arch 3.5 or greater. + if (${EIGEN_CUDA_COMPUTE_ARCH} GREATER 34) + ei_add_test(cxx11_tensor_random_cuda) + endif() + + # Operations other that casting of half floats are only supported starting with arch 5.3 if (${EIGEN_CUDA_COMPUTE_ARCH} GREATER 52) ei_add_test(cxx11_tensor_of_float16_cuda) endif() |