| /external/tensorflow/tensorflow/examples/adding_an_op/ |
| cuda_op_kernel.cu.cc | 21 for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < N;
22 i += blockDim.x * gridDim.x) {
|
| /external/tensorflow/tensorflow/tools/ci_build/builds/user_ops/ |
| cuda_op_kernel.cu.cc | 21 for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < N;
22 i += blockDim.x * gridDim.x) {
|
| /external/tensorflow/tensorflow/core/kernels/ |
| concat_lib_gpu_impl.cu.cc | 40 IntType gidx = blockIdx.x * blockDim.x + threadIdx.x;
42 for (; gidx < total_cols; gidx += blockDim.x * gridDim.x) {
43 IntType gidy = blockIdx.y * blockDim.y + threadIdx.y;
49 for (; gidy < total_rows; gidy += blockDim.y * gridDim.y) {
68 IntType gidx = blockIdx.x * blockDim.x + threadIdx.x;
76 IntType lidx = threadIdx.y * blockDim.x + threadIdx.x;
77 IntType blockSize = blockDim.x * blockDim.y;
96 for (; gidx < total_cols; gidx += blockDim.x * gridDim.x) {
107 IntType gidy = blockIdx.y * blockDim.y + threadIdx.y
[all...] |
| split_lib_gpu.cu.cc | 78 eigen_assert(blockDim.y == 1);
79 eigen_assert(blockDim.z == 1);
118 IntType gidx = blockIdx.x * blockDim.x + threadIdx.x;
126 IntType lidx = threadIdx.y * blockDim.x + threadIdx.x;
127 IntType blockSize = blockDim.x * blockDim.y;
146 for (; gidx < total_cols; gidx += blockDim.x * gridDim.x) {
157 IntType gidy = blockIdx.y * blockDim.y + threadIdx.y;
158 for (; gidy < total_rows; gidy += blockDim.y * gridDim.y)
173 eigen_assert(blockDim.y == 1)
[all...] |
| check_numerics_op_gpu.cu.cc | 39 const int32 thread_id = blockIdx.x * blockDim.x + threadIdx.x;
40 const int32 total_thread_count = gridDim.x * blockDim.x;
|
| bucketize_op_gpu.cu.cc | 46 int32 lidx = threadIdx.y * blockDim.x + threadIdx.x;
47 int32 blockSize = blockDim.x * blockDim.y;
|
| bias_op_gpu.cu.cc | 121 for (int32 index = threadIdx.x; index < bias_size; index += blockDim.x) {
126 for (int32 index = blockIdx.x * blockDim.x + threadIdx.x; index < nthreads;
127 index += blockDim.x * gridDim.x) {
133 for (int32 index = threadIdx.x; index < bias_size; index += blockDim.x) {
147 for (int32 index = threadIdx.x; index < kSDataSize; index += blockDim.x) {
158 for (int32 index = group_index * blockDim.x + threadIdx.x;
159 index < total_count; index += blockDim.x * group_size) {
|
| random_op_gpu.cu.cc | 134 const int32 thread_id = blockIdx.x * blockDim.x + threadIdx.x;
135 const int32 total_thread_count = gridDim.x * blockDim.x;
175 const int32 thread_id = blockIdx.x * blockDim.x + threadIdx.x;
176 const int32 total_thread_count = gridDim.x * blockDim.x;
|
| /external/eigen/unsupported/Eigen/CXX11/src/Tensor/ |
| TensorReductionCuda.h | 115 const Index thread_id = blockIdx.x * blockDim.x + threadIdx.x;
116 const Index num_threads = blockDim.x * gridDim.x;
192 eigen_assert(blockDim.x == 1);
205 const Index thread_id = blockIdx.x * blockDim.x + threadIdx.x;
206 const Index num_threads = blockDim.x * gridDim.x;
378 eigen_assert(blockDim.y == 1);
379 eigen_assert(blockDim.z == 1);
386 const Index input_col_blocks = divup<Index>(num_coeffs_to_reduce, blockDim.x * NumPerThread);
389 const Index num_threads = blockDim.x * gridDim.x;
390 const Index thread_id = blockIdx.x * blockDim.x + threadIdx.x
[all...] |
| /external/tensorflow/tensorflow/contrib/mpi_collectives/kernels/ |
| ring.cu.cc | 90 for (size_t i = blockIdx.x * blockDim.x + threadIdx.x; i < N;
91 i += blockDim.x * gridDim.x) {
|
| /external/tensorflow/tensorflow/core/util/ |
| cuda_device_functions.h | 90 return detail::CudaGridRange<T>(blockIdx.x * blockDim.x + threadIdx.x,
91 gridDim.x * blockDim.x, count);
98 return detail::CudaGridRange<T>(blockIdx.y * blockDim.y + threadIdx.y,
99 gridDim.y * blockDim.y, count);
106 return detail::CudaGridRange<T>(blockIdx.z * blockDim.z + threadIdx.z,
107 gridDim.z * blockDim.z, count);
386 assert(blockDim.y == 1 && blockDim.z == 1);
387 assert(blockDim.x * gridDim.x / blockDim.x == gridDim.x)
[all...] |
| /external/eigen/test/ |
| cuda_common.h | 11 dim3 threadIdx, blockDim, blockIdx;
26 int i = threadIdx.x + blockIdx.x*blockDim.x;
|
| /external/clang/lib/Headers/ |
| cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/darwin-x86/clang-3289846/lib64/clang/3.8/include/ |
| cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/darwin-x86/clang-3859424/lib64/clang/4.0/include/ |
| __clang_cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/darwin-x86/clang-4053586/lib64/clang/5.0/include/ |
| __clang_cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/darwin-x86/clang-4393122/lib64/clang/5.0.1/include/ |
| __clang_cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/darwin-x86/clang-4479392/lib64/clang/5.0.2/include/ |
| __clang_cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/darwin-x86/clang-4579689/lib64/clang/6.0.1/include/ |
| __clang_cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/darwin-x86/clang-4630689/lib64/clang/6.0.1/include/ |
| __clang_cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/darwin-x86/clang-4639204/lib64/clang/6.0.1/include/ |
| __clang_cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/darwin-x86/clang-4691093/lib64/clang/6.0.2/include/ |
| __clang_cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/linux-x86/clang-3289846/lib64/clang/3.8/include/ |
| cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/linux-x86/clang-3859424/lib64/clang/4.0/include/ |
| __clang_cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
| /prebuilts/clang/host/linux-x86/clang-4053586/lib64/clang/5.0/include/ |
| __clang_cuda_builtin_vars.h | 92 // blockDim should be convertible to dim3 (in fact in nvcc, it *is* a
114 __CUDA_BUILTIN_VAR __cuda_builtin_blockDim_t blockDim;
|
