| /external/tensorflow/tensorflow/contrib/reduce_slice_ops/kernels/ |
| D | reduce_slice_ops_gpu.cu.cc | 42 CUDA_AXIS_KERNEL_LOOP(x, config.virtual_thread_count.x, X) { \
43 CUDA_AXIS_KERNEL_LOOP(y, config.virtual_thread_count.y, Y) { \
44 CUDA_AXIS_KERNEL_LOOP(z, config.virtual_thread_count.z, Z) { \
45 Index outidx = x * config.virtual_thread_count.y * \
46 config.virtual_thread_count.z + \
47 y * config.virtual_thread_count.z + z; \
52 Index inidx = x * bound * config.virtual_thread_count.z + \
53 yin * config.virtual_thread_count.z + z; \
|
| /external/tensorflow/tensorflow/core/util/ |
| D | cuda_launch_config.h | 114 int virtual_thread_count = -1; member
129 const int virtual_thread_count = work_element_count; in GetCudaLaunchConfig() local
132 virtual_thread_count); in GetCudaLaunchConfig()
138 config.virtual_thread_count = virtual_thread_count; in GetCudaLaunchConfig()
166 config.virtual_thread_count = work_element_count; in GetCudaLaunchConfig()
190 config.virtual_thread_count = work_element_count; in GetCudaLaunchConfigFixedBlockSize()
197 dim3 virtual_thread_count = dim3(0, 0, 0); member
220 config.virtual_thread_count = dim3(xdim, ydim, 1); in GetCuda2DLaunchConfig()
276 config.virtual_thread_count = dim3(xdim, ydim, zdim); in GetCuda3DLaunchConfig()
|
| D | cuda_kernel_helper_test.cu.cc | 44 CUDA_1D_KERNEL_LOOP(x, config.virtual_thread_count) { outbuf[x] = 0; } in SetOutbufZero()
49 CUDA_1D_KERNEL_LOOP(x, config.virtual_thread_count) { in Count1D()
57 CUDA_AXIS_KERNEL_LOOP(x, config.virtual_thread_count.x, X) { in Count2D()
61 CUDA_AXIS_KERNEL_LOOP(y, config.virtual_thread_count.y, Y) { in Count2D()
65 int idx = x * config.virtual_thread_count.y + y; in Count2D()
71 CUDA_AXIS_KERNEL_LOOP(x, config.virtual_thread_count.x, X) { in Count3D()
75 CUDA_AXIS_KERNEL_LOOP(y, config.virtual_thread_count.y, Y) { in Count3D()
79 CUDA_AXIS_KERNEL_LOOP(z, config.virtual_thread_count.z, Z) { in Count3D()
84 x * config.virtual_thread_count.y * config.virtual_thread_count.z + in Count3D()
85 y * config.virtual_thread_count.z + z; in Count3D()
|
| /external/tensorflow/tensorflow/core/kernels/ |
| D | diag_op_gpu.cu.cc | 58 int virtual_thread_count = int(size * size); in operator ()() local
63 GetCudaLaunchConfig(virtual_thread_count, device); in operator ()()
65 device.stream()>>>(diag_config.virtual_thread_count, size, in operator ()()
105 0, device.stream()>>>(diag_config.virtual_thread_count, in operator ()()
|
| D | inplace_ops_functor_gpu.cu.cc | 54 d.stream(), cfg.virtual_thread_count, nrows, ncols, loc, src, dst)); in DoParallelConcatUpdate()
122 d.stream(), cfg.virtual_thread_count, nrows, in DoInplaceOp()
128 cfg.virtual_thread_count, nrows, ncols, n, in DoInplaceOp()
134 cfg.virtual_thread_count, nrows, ncols, n, in DoInplaceOp()
156 d.stream(), cfg.virtual_thread_count, nrows, in DoInplaceOp()
|
| D | resize_bilinear_op_gpu.cu.cc | 286 0, d.stream(), config.virtual_thread_count, images.data(), in operator ()()
292 config.thread_per_block, 0, d.stream(), config.virtual_thread_count, in operator ()()
324 config.virtual_thread_count, output_grad.data())); in operator ()()
332 config.thread_per_block, 0, d.stream(), config.virtual_thread_count, in operator ()()
339 config.thread_per_block, 0, d.stream(), config.virtual_thread_count, in operator ()()
|
| D | fused_batch_norm_op.cu.cc | 42 d.stream()>>>(config.virtual_thread_count, in operator ()()
64 d.stream()>>>(config.virtual_thread_count, in operator ()()
|
| D | matrix_set_diag_op_gpu.cu.cc | 79 config.virtual_thread_count, m, n, minsize, in Compute()
86 config.virtual_thread_count, m, n, minsize, in Compute()
|
| D | crop_and_resize_op_gpu.cu.cc | 376 0, d.stream(), config.virtual_thread_count, image.data(), in operator ()()
411 d.stream(), config.virtual_thread_count, grads_image.data())); in operator ()()
426 config.thread_per_block, 0, d.stream(), config.virtual_thread_count, in operator ()()
461 d.stream(), config.virtual_thread_count, grads_boxes.data())); in operator ()()
470 config.thread_per_block, 0, d.stream(), config.virtual_thread_count, in operator ()()
|
| D | determinant_op_gpu.cu.cc | 136 config.virtual_thread_count, n, lu_factor.data(), pivots, nullptr, in operator ()()
158 config.virtual_thread_count, n, lu_factor.data(), pivots, sign.data(), in operator ()()
|
| D | svd_op_gpu.cu.cc | 66 CUDA_AXIS_KERNEL_LOOP(batch, config.virtual_thread_count.x, X) { in ComputeValueOfVKernel()
67 CUDA_AXIS_KERNEL_LOOP(i, config.virtual_thread_count.y, Y) { in ComputeValueOfVKernel()
78 CUDA_1D_KERNEL_LOOP(i, config.virtual_thread_count) { in ExtractSignOfVKernel()
|
| D | dilation_ops_gpu.cu.cc | 201 config.virtual_thread_count, input.data(), filter.data(), batch, in operator ()()
242 config.thread_per_block, 0, d.stream(), config.virtual_thread_count, in operator ()()
284 config.virtual_thread_count, input.data(), filter.data(), in operator ()()
|
| D | conv_2d_gpu.h | 442 d.stream(), config.virtual_thread_count,
464 d.stream(), config.virtual_thread_count,
495 config.thread_per_block, 0, d.stream(), config.virtual_thread_count,
500 config.thread_per_block, 0, d.stream(), config.virtual_thread_count,
936 d.stream(), config.virtual_thread_count, input,
969 d.stream(), config.virtual_thread_count, in,
|
| D | depthtospace_op_gpu.cu.cc | 167 config.virtual_thread_count, input.data(), block_size, batch_size, in operator ()()
232 config.virtual_thread_count, input.data(), block_size, input_width, in operator ()()
|
| D | eye_functor_gpu.cu.cc | 56 device.stream()>>>(config.virtual_thread_count, batch_size, m, in operator ()()
|
| D | spacetodepth_op_gpu.cu.cc | 163 config.virtual_thread_count, input.data(), block_size, batch_size, in operator ()()
228 config.virtual_thread_count, input.data(), block_size, output_width, in operator ()()
|
| D | matrix_band_part_op_gpu.cu.cc | 63 config.virtual_thread_count, batch_size, m, n, num_lower_diags, in operator ()()
|
| D | resize_nearest_neighbor_op_gpu.cu.cc | 235 input_config.virtual_thread_count, input.data(), in_height, in_width, in operator ()()
243 d.stream(), input_config.virtual_thread_count, input.data(), in operator ()()
|
| D | bias_op_gpu.cu.cc | 90 config.virtual_thread_count, input, bias, in compute()
95 config.virtual_thread_count, input, bias, in compute()
|
| D | tile_functor_gpu.h | 80 d.stream(), cfg.virtual_thread_count, p, in TileSimple()
|
| D | avgpooling_op_gpu.cu.cc | 96 d.stream(), config.virtual_thread_count, top_diff, num, height, width, in RunAvePoolBackwardNHWC()
|
| D | multinomial_op_gpu.cu.cc | 115 d.stream()>>>(config.virtual_thread_count, num_classes, in operator ()()
|
| D | spacetobatch_functor_gpu.cu.cc | 146 config.virtual_thread_count, in operator ()()
|
| D | lu_op_gpu.cu.cc | 69 CUDA_1D_KERNEL_LOOP(index, config.virtual_thread_count) { in ComputePermutationFromTranspositionsKernel()
|
| D | transpose_functor_gpu.cu.cc | 85 d.stream(), cfg.virtual_thread_count, p, in TransposeSimple()
|