/external/tensorflow/tensorflow/core/kernels/ |
D | population_count_op_gpu.cu.cc | 39 CUDA_1D_KERNEL_LOOP(i, size) { output[i] = __popc(ldg(input + i)); } in PopulationCountKernel() 46 CUDA_1D_KERNEL_LOOP(i, size) { in PopulationCountKernel() 55 CUDA_1D_KERNEL_LOOP(i, size) { in PopulationCountKernel() 63 CUDA_1D_KERNEL_LOOP(i, size) { output[i] = __popcll(ldg(input + i)); } in PopulationCountKernel()
|
D | compare_and_bitpack_op_gpu.cu.cc | 45 CUDA_1D_KERNEL_LOOP(i, size) { in CompareAndBitpackKernel() 65 CUDA_1D_KERNEL_LOOP(i, size) { in CompareAndBitpackKernel() 85 CUDA_1D_KERNEL_LOOP(i, size) { in CompareAndBitpackKernel() 102 CUDA_1D_KERNEL_LOOP(i, size) { in CompareAndBitpackKernel()
|
D | maxpooling_op_gpu.cu.cc | 72 CUDA_1D_KERNEL_LOOP(index, nthreads) { in MaxPoolForwardNCHW() 116 CUDA_1D_KERNEL_LOOP(index, nthreads) { in MaxPoolForwardNoMaskKernel_NCHW_VECT_C() 146 CUDA_1D_KERNEL_LOOP(index, nthreads) { in MaxPoolForwardNHWC() 185 CUDA_1D_KERNEL_LOOP(index, nthreads) { in MaxPoolBackwardNoMaskNHWC() 244 CUDA_1D_KERNEL_LOOP(index, nthreads) { in MaxPoolBackward() 272 CUDA_1D_KERNEL_LOOP(index, nthreads) { in MaxPoolGradBackwardNoMaskNCHW() 312 CUDA_1D_KERNEL_LOOP(index, nthreads) { in MaxPoolGradBackwardNoMaskNHWC() 371 CUDA_1D_KERNEL_LOOP(index, nthreads) { in MaxPoolGradBackward() 378 #undef CUDA_1D_KERNEL_LOOP
|
D | cwise_op_clip_gpu.cu.cc | 29 CUDA_1D_KERNEL_LOOP(i, size_in) { in UnaryClipCustomKernel() 39 CUDA_1D_KERNEL_LOOP(i, size_in) { in BinaryRightClipCustomKernel() 48 CUDA_1D_KERNEL_LOOP(i, size_in) { in BinaryLeftClipCustomKernel()
|
D | resize_nearest_neighbor_op_gpu.cu.cc | 41 CUDA_1D_KERNEL_LOOP(index, nthreads) { in ResizeNearestNeighborNHWC() 72 CUDA_1D_KERNEL_LOOP(index, nthreads) { in LegacyResizeNearestNeighborNHWC() 101 CUDA_1D_KERNEL_LOOP(index, nthreads) { in ResizeNearestNeighborBackwardNHWC() 132 CUDA_1D_KERNEL_LOOP(index, nthreads) { in LegacyResizeNearestNeighborBackwardNHWC()
|
D | fused_batch_norm_op.cu.cc | 31 CUDA_1D_KERNEL_LOOP(index, nthreads) { in VarianceToInvVarianceKernel() 49 CUDA_1D_KERNEL_LOOP(index, nthreads) { in InvVarianceToVarianceKernel()
|
D | bias_op_gpu.cu.cc | 58 CUDA_1D_KERNEL_LOOP(index, nthreads) { in BiasNHWCKernel() 67 CUDA_1D_KERNEL_LOOP(index, nthreads) { in BiasNCHWKernel() 104 CUDA_1D_KERNEL_LOOP(index, nthreads) { in BiasGradNHWC_Naive() 115 CUDA_1D_KERNEL_LOOP(index, nthreads) { in BiasGradNCHW_Naive()
|
D | matrix_set_diag_op_gpu.cu.cc | 34 CUDA_1D_KERNEL_LOOP(index, num_threads) { in MatrixSetDiagKernel() 46 CUDA_1D_KERNEL_LOOP(index, num_threads) { in MatrixCopyInputAndSetDiagKernel()
|
D | diag_op_gpu.cu.cc | 33 CUDA_1D_KERNEL_LOOP(index, num_threads) { in DiagCudaKernel() 87 CUDA_1D_KERNEL_LOOP(index, num_threads) { in DiagPartCudaKernel()
|
D | resize_bilinear_op_gpu.cu.cc | 41 CUDA_1D_KERNEL_LOOP(out_idx, nthreads) { in ResizeBilinearKernel() 92 CUDA_1D_KERNEL_LOOP(in_idx, nthreads) { in ResizeBilinearGradKernel() 156 CUDA_1D_KERNEL_LOOP(out_idx, nthreads) { in LegacyResizeBilinearKernel() 206 CUDA_1D_KERNEL_LOOP(in_idx, nthreads) { in LegacyResizeBilinearGradKernel()
|
D | depthtospace_op_gpu.cu.cc | 41 CUDA_1D_KERNEL_LOOP(out_idx, nthreads) { in D2S_NHWC() 70 CUDA_1D_KERNEL_LOOP(input_idx, nthreads) { in D2S_NCHW() 108 CUDA_1D_KERNEL_LOOP(thread_idx, nthreads) { in D2S_NCHW_LOOP()
|
D | spacetodepth_op_gpu.cu.cc | 39 CUDA_1D_KERNEL_LOOP(inp_idx, nthreads) { in S2D_NHWC() 69 CUDA_1D_KERNEL_LOOP(input_idx, nthreads) { in S2D_NCHW() 108 CUDA_1D_KERNEL_LOOP(thread_idx, nthreads) { in S2D_NCHW_LOOP()
|
D | searchsorted_op_gpu.cu.cc | 38 CUDA_1D_KERNEL_LOOP(work_unit_id, values_size * batch_size) { in UpperBoundKernel() 50 CUDA_1D_KERNEL_LOOP(work_unit_id, values_size * batch_size) { in LowerBoundKernel()
|
D | dilation_ops_gpu.cu.cc | 46 CUDA_1D_KERNEL_LOOP(out_idx, nthreads) { in DilationKernel() 85 CUDA_1D_KERNEL_LOOP(out_idx, nthreads) { in DilationBackpropInputKernel() 134 CUDA_1D_KERNEL_LOOP(out_idx, nthreads) { in DilationBackpropFilterKernel()
|
D | scatter_functor_gpu.cu.h | 79 CUDA_1D_KERNEL_LOOP(i, updates_size) { 100 CUDA_1D_KERNEL_LOOP(i, synthesized_updates_size) {
|
D | pooling_ops_3d_gpu.cu.cc | 38 CUDA_1D_KERNEL_LOOP(index, nthreads) { in MaxPoolGradBackwardNoMaskNCDHW() 88 CUDA_1D_KERNEL_LOOP(index, nthreads) { in MaxPoolGradBackwardNoMaskNDHWC()
|
D | eye_functor_gpu.cu.cc | 36 CUDA_1D_KERNEL_LOOP(index, num_threads) { in EyeKernel()
|
D | inplace_ops_functor_gpu.cu.cc | 33 CUDA_1D_KERNEL_LOOP(idx, nthreads) { in DoParallelConcatOpKernel() 85 CUDA_1D_KERNEL_LOOP(idx, nthreads) { in DoInplaceOpKernel()
|
D | matrix_band_part_op_gpu.cu.cc | 38 CUDA_1D_KERNEL_LOOP(index, num_threads) { in MatrixBandPartKernel()
|
D | crop_and_resize_op_gpu.cu.cc | 46 CUDA_1D_KERNEL_LOOP(out_idx, nthreads) { in CropAndResizeKernel() 138 CUDA_1D_KERNEL_LOOP(out_idx, nthreads) { in CropAndResizeBackpropImageKernel() 233 CUDA_1D_KERNEL_LOOP(out_idx, nthreads) { in CropAndResizeBackpropBoxesKernel()
|
D | dynamic_stitch_op_gpu.cu.cc | 39 CUDA_1D_KERNEL_LOOP(output_index, output_size) { in DynamicStitchKernel()
|
D | split_lib_gpu.cu.cc | 91 CUDA_1D_KERNEL_LOOP(offset, size) { in SplitOpKernel() 185 CUDA_1D_KERNEL_LOOP(offset, size) { in SplitVOpKernel_fixed()
|
D | tile_functor_gpu.h | 39 CUDA_1D_KERNEL_LOOP(o_idx, nthreads) { in TileKernel()
|
/external/tensorflow/tensorflow/contrib/resampler/kernels/ |
D | resampler_ops_gpu.cu.cc | 46 CUDA_1D_KERNEL_LOOP(index, output_data_size) { in Resampler2DKernel() 152 CUDA_1D_KERNEL_LOOP(index, resampler_output_size) { in ResamplerGrad2DKernel()
|
/external/tensorflow/tensorflow/core/util/ |
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()
|