diff options
author | Benoit Steiner <benoit.steiner.goog@gmail.com> | 2015-07-15 12:38:34 -0700 |
---|---|---|
committer | Benoit Steiner <benoit.steiner.goog@gmail.com> | 2015-07-15 12:38:34 -0700 |
commit | e892524efe7e8adbd43bf4c1c150f4f4ebf27d1d (patch) | |
tree | 4bc0c43b27ec1ee6a595baf0e56904da5b309855 /unsupported/Eigen/CXX11/src/Tensor/TensorConvolution.h | |
parent | f5aa64086228ca9ccfa27e6086667fd0bdbad22c (diff) |
Added support for multi gpu configuration to the GpuDevice class
Diffstat (limited to 'unsupported/Eigen/CXX11/src/Tensor/TensorConvolution.h')
-rw-r--r-- | unsupported/Eigen/CXX11/src/Tensor/TensorConvolution.h | 8 |
1 files changed, 4 insertions, 4 deletions
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorConvolution.h b/unsupported/Eigen/CXX11/src/Tensor/TensorConvolution.h index 88db9d410..64b0315ee 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorConvolution.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorConvolution.h @@ -835,10 +835,10 @@ struct TensorEvaluator<const TensorConvolutionOp<Indices, InputArgType, KernelAr void executeEval(Scalar* data) const { typedef typename TensorEvaluator<InputArgType, GpuDevice>::Dimensions InputDims; - const int maxSharedMem = sharedMemPerBlock(); - const int maxThreadsPerBlock = maxCudaThreadsPerBlock(); - const int maxBlocksPerProcessor = maxCudaThreadsPerMultiProcessor() / maxThreadsPerBlock; - const int numMultiProcessors = getNumCudaMultiProcessors(); + const int maxSharedMem = m_device.sharedMemPerBlock(); + const int maxThreadsPerBlock = m_device.maxCudaThreadsPerBlock(); + const int maxBlocksPerProcessor = m_device.maxCudaThreadsPerMultiProcessor() / maxThreadsPerBlock; + const int numMultiProcessors = m_device.getNumCudaMultiProcessors(); const int warpSize = 32; switch (NumKernelDims) { |