/aosp_15_r20/external/pytorch/aten/src/ATen/native/cuda/ |
H A D | Sorting.cu | 253 cuda::detail::canUse32BitIndexMath(self) && in launch_kthvalue_kernel() 254 cuda::detail::canUse32BitIndexMath(values) && in launch_kthvalue_kernel() 255 cuda::detail::canUse32BitIndexMath(indices) ? ScalarType::Int : ScalarType::Long, in launch_kthvalue_kernel() 268 if (cuda::detail::canUse32BitIndexMath(vals) && in launch_median_kernel() 269 cuda::detail::canUse32BitIndexMath(inds) && in launch_median_kernel() 270 cuda::detail::canUse32BitIndexMath(self)) { in launch_median_kernel()
|
H A D | ReflectionPad.cu | 31 using at::cuda::detail::canUse32BitIndexMath; 274 TORCH_CHECK(canUse32BitIndexMath(input_), in reflection_pad2d_out_template() 364 TORCH_CHECK(canUse32BitIndexMath(input), in reflection_pad2d_backward_out_template() 366 TORCH_CHECK(canUse32BitIndexMath(grad_output_), in reflection_pad2d_backward_out_template() 433 canUse32BitIndexMath(input_), in TORCH_IMPL_FUNC() 491 TORCH_CHECK(canUse32BitIndexMath(input), in TORCH_IMPL_FUNC() 494 TORCH_CHECK(canUse32BitIndexMath(grad_output_), in TORCH_IMPL_FUNC() 574 canUse32BitIndexMath(input_), in TORCH_IMPL_FUNC() 628 TORCH_CHECK(canUse32BitIndexMath(input), "input tensor must fit into 32-bit index math"); in TORCH_IMPL_FUNC() 629 TORCH_CHECK(canUse32BitIndexMath(grad_output), "input tensor must fit into 32-bit index math"); in TORCH_IMPL_FUNC()
|
H A D | GridSampler.cu | 766 if (canUse32BitIndexMath(input) && canUse32BitIndexMath(grid) && in launch_grid_sampler_2d_forward_kernel() 767 canUse32BitIndexMath(output)) { in launch_grid_sampler_2d_forward_kernel() 811 if (canUse32BitIndexMath(input) && canUse32BitIndexMath(grid) && in launch_grid_sampler_3d_forward_kernel() 812 canUse32BitIndexMath(output)) { in launch_grid_sampler_3d_forward_kernel() 866 if (canUse32BitIndexMath(input) && canUse32BitIndexMath(grid) && in launch_grid_sampler_2d_backward_kernel() 867 canUse32BitIndexMath(grad_output)) { in launch_grid_sampler_2d_backward_kernel() 925 if (canUse32BitIndexMath(input) && canUse32BitIndexMath(grid) && in launch_grid_sampler_3d_backward_kernel() 926 canUse32BitIndexMath(grad_output)) { in launch_grid_sampler_3d_backward_kernel()
|
H A D | Normalization.cu | 73 if (!at::cuda::detail::canUse32BitIndexMath(self)) { in batch_norm_choose_impl() 583 cuda::detail::canUse32BitIndexMath(input) && in batch_norm_backward_cuda() 584 cuda::detail::canUse32BitIndexMath(grad_out)) { in batch_norm_backward_cuda() 659 if (cuda::detail::canUse32BitIndexMath(self)) { in batch_norm_stats_cuda() 717 if (cuda::detail::canUse32BitIndexMath(self)) { in batch_norm_gather_stats_with_counts_cuda() 730 if (at::cuda::detail::canUse32BitIndexMath(grad_output) && in batch_norm_backward_reduce_cuda() 746 if (cuda::detail::canUse32BitIndexMath(grad_output)) { in batch_norm_backward_reduce_cuda() 767 if (at::cuda::detail::canUse32BitIndexMath(self) && in batch_norm_backward_elemt_cuda() 780 if (cuda::detail::canUse32BitIndexMath(self)) { in batch_norm_backward_elemt_cuda()
|
H A D | DepthwiseConv2d.cu | 407 TORCH_CHECK(canUse32BitIndexMath(input) && canUse32BitIndexMath(output)); in conv_depthwise2d_forward_out() 494 TORCH_CHECK(canUse32BitIndexMath(grad_input) && in conv_depthwise2d_backward_out() 495 canUse32BitIndexMath(grad_output)); in conv_depthwise2d_backward_out() 622 TORCH_CHECK(canUse32BitIndexMath(input) && in conv_depthwise2d_grad_weight_out() 623 canUse32BitIndexMath(grad_output)); in conv_depthwise2d_grad_weight_out()
|
H A D | ReplicationPadding.cu | 235 TORCH_CHECK(at::cuda::detail::canUse32BitIndexMath(input), in replication_pad2d_backward_out_cuda_template() 237 TORCH_CHECK(at::cuda::detail::canUse32BitIndexMath(gradOutput), in replication_pad2d_backward_out_cuda_template() 310 TORCH_CHECK(at::cuda::detail::canUse32BitIndexMath(input), in shapeAndGradOutputCheck3d() 344 TORCH_CHECK(at::cuda::detail::canUse32BitIndexMath(gradOutput), in shapeAndGradOutputCheck3d() 537 TORCH_CHECK(at::cuda::detail::canUse32BitIndexMath(input), in TORCH_IMPL_FUNC()
|
H A D | Indexing.cu | 918 if (cuda::detail::canUse32BitIndexMath(result) && in index_add_cuda_impl() 919 cuda::detail::canUse32BitIndexMath(source) && in index_add_cuda_impl() 920 cuda::detail::canUse32BitIndexMath(index)) { in index_add_cuda_impl() 1091 if (cuda::detail::canUse32BitIndexMath(result) && in index_reduce_func_cuda_impl() 1092 cuda::detail::canUse32BitIndexMath(source) && in index_reduce_func_cuda_impl() 1093 cuda::detail::canUse32BitIndexMath(index)) { in index_reduce_func_cuda_impl() 1387 if (cuda::detail::canUse32BitIndexMath(out) && in index_select_out_cuda_impl() 1388 cuda::detail::canUse32BitIndexMath(self) && in index_select_out_cuda_impl() 1389 cuda::detail::canUse32BitIndexMath(index)) { in index_select_out_cuda_impl()
|
H A D | TriangularOps.cu | 125 if (cuda::detail::canUse32BitIndexMath(result) && cuda::detail::canUse32BitIndexMath(self)) { in triu_tril_cuda_template()
|
H A D | RNN.cu | 29 using at::cuda::detail::canUse32BitIndexMath; 539 if (canUse32BitIndexMath(workspace)) { // See Note [64-bit index math check elision] in _thnn_fused_lstm_cell_cuda() 590 if (canUse32BitIndexMath(workspace)) { // See Note [64-bit index math check elision] in _thnn_fused_lstm_cell_backward_impl_cuda() 624 if (canUse32BitIndexMath(workspace)) { // See Note [64-bit index math check elision] in _thnn_fused_gru_cell_cuda() 653 if (canUse32BitIndexMath(workspace)) { // See Note [64-bit index math check elision] in _thnn_fused_gru_cell_backward_cuda()
|
H A D | Shape.cu | 478 return at::cuda::detail::canUse32BitIndexMath(t); in TORCH_IMPL_FUNC() 493 at::cuda::detail::canUse32BitIndexMath(result) && in TORCH_IMPL_FUNC() 510 at::cuda::detail::canUse32BitIndexMath(result) && in TORCH_IMPL_FUNC()
|
H A D | TensorModeKernel.cpp | 79 canUse32BitIndexMath(self)) { in mode_kernel_impl()
|
H A D | TensorTopK.cu | 894 if (at::cuda::detail::canUse32BitIndexMath(input) && in launch_gather_topk_kernel() 895 at::cuda::detail::canUse32BitIndexMath(values) && in launch_gather_topk_kernel() 896 at::cuda::detail::canUse32BitIndexMath(indices)) { in launch_gather_topk_kernel()
|
H A D | IndexKernel.cu | 326 …AT_DISPATCH_INDEX_TYPES(cuda::detail::canUse32BitIndexMath(output) ? ScalarType::Int : ScalarType:… in put_kernel() 351 …AT_DISPATCH_INDEX_TYPES(cuda::detail::canUse32BitIndexMath(input) ? ScalarType::Int : ScalarType::… in take_kernel()
|
H A D | Sort.cu | 311 if (at::cuda::detail::canUse32BitIndexMath(key)) { in sortCommon()
|
H A D | Dropout.cu | 358 if (cuda::detail::canUse32BitIndexMath(self)){ in dropout_cuda()
|
H A D | NLLLoss2d.cu | 325 at::native::canUse32BitIndexMath(input_, INT_MAX) ? ScalarType::Int : ScalarType::Long, in nll_loss2d_forward_out_cuda_template()
|
H A D | AveragePool2d.cu | 417 at::native::canUse32BitIndexMath(input, INT_MAX) ? ScalarType::Int : ScalarType::Long, in TORCH_IMPL_FUNC()
|
H A D | SoftMax.cu | 929 at::native::canUse32BitIndexMath(input, INT_MAX) ? ScalarType::Int : ScalarType::Long, in host_softmax()
|
/aosp_15_r20/external/pytorch/aten/src/ATen/native/ |
H A D | GridSamplerUtils.h | 99 at::native::canUse32BitIndexMath(input) && in cond_cudnn_grid_sampler() 100 at::native::canUse32BitIndexMath(grid) && in cond_cudnn_grid_sampler()
|
H A D | CanUse32BitIndexMath.h | 11 TORCH_API bool canUse32BitIndexMath(const at::TensorBase &t, int64_t max_elem=std::numeric_limits<i…
|
H A D | IndexingUtils.cpp | 6 bool canUse32BitIndexMath(const TensorBase& t, int64_t max_elem) { in canUse32BitIndexMath() function
|
/aosp_15_r20/external/pytorch/aten/src/ATen/cuda/ |
H A D | CUDAApplyUtils.cuh | 477 if (detail::canUse32BitIndexMath(a) && in CUDA_tensor_apply2() 478 detail::canUse32BitIndexMath(b)) { in CUDA_tensor_apply2()
|
/aosp_15_r20/external/pytorch/aten/src/ATen/cuda/detail/ |
H A D | IndexUtils.cuh | 10 using at::native::canUse32BitIndexMath;
|