| /aosp_15_r20/external/pytorch/aten/src/ATen/native/cuda/ |
| H A D | Reduce.cuh | 1157 bool can_use_32bit_indexing = iter.can_use_32bit_indexing(); in gpu_reduce_kernel() local
1164 if (!can_accumulate_in_output && !can_use_32bit_indexing) { in gpu_reduce_kernel()
1180 if (!can_use_32bit_indexing) { in gpu_reduce_kernel()
1213 AT_ASSERT(can_use_32bit_indexing); in gpu_reduce_kernel()
1264 bool can_use_32bit_indexing = iter.can_use_32bit_indexing(); in jitted_gpu_reduce_kernel() local
1272 if (!can_accumulate_in_output && !can_use_32bit_indexing) { in jitted_gpu_reduce_kernel()
1288 if (!can_use_32bit_indexing) { in jitted_gpu_reduce_kernel()
1323 AT_ASSERT(can_use_32bit_indexing); in jitted_gpu_reduce_kernel()
|
| H A D | IndexKernel.cu | 64 if (!iter.can_use_32bit_indexing()) { in gpu_index_kernel()
119 if (!iter.can_use_32bit_indexing()) { in index_fill_kernel_impl()
156 if (!iter.can_use_32bit_indexing()) { in index_copy_kernel_impl()
280 if (!iter.can_use_32bit_indexing()) { in cuda_take_put_kernel()
430 if (!iter.can_use_32bit_indexing()) { in flip_kernel_impl()
|
| H A D | Loops.cuh | 88 if (!iter.can_use_32bit_indexing()) { in gpu_kernel_nocast()
111 if (!iter.can_use_32bit_indexing()) { in gpu_kernel()
282 TORCH_INTERNAL_ASSERT(iter.can_use_32bit_indexing()); in gpu_kernel_multiple_outputs_impl()
316 if (!iter.can_use_32bit_indexing()) { in gpu_kernel_multiple_outputs()
|
| H A D | CUDALoops.cuh | 256 TORCH_INTERNAL_ASSERT(iter.can_use_32bit_indexing()); in gpu_kernel_impl_nocast()
291 TORCH_INTERNAL_ASSERT(iter.can_use_32bit_indexing()); in gpu_kernel_impl()
|
| H A D | DistributionTemplates.h | 137 if (!iter.can_use_32bit_indexing()) { in distribution_nullary_kernel()
241 if (!iter.can_use_32bit_indexing()) { in distribution_binary_kernel()
248 TORCH_INTERNAL_ASSERT_DEBUG_ONLY(iter.can_use_32bit_indexing()); in distribution_binary_kernel()
|
| H A D | Activation.cpp | 60 if (iter.can_use_32bit_indexing()) { in glu_backward_cuda_out()
|
| H A D | ScatterGatherKernel.cu | 126 if (!iter.can_use_32bit_indexing()) { in operator ()()
348 if (!iter.can_use_32bit_indexing()) { in operator ()()
|
| H A D | FunctionOfAMatrixUtilsKernel.cu | 56 if (!iter.can_use_32bit_indexing()) { in _compute_linear_combination_internal_kernel()
|
| H A D | CrossKernel.cu | 81 if (iter.can_use_32bit_indexing()) { in cross_impl()
|
| H A D | JitLoops.cuh | 88 if (!iter.can_use_32bit_indexing()) { in jitted_gpu_kernel()
|
| H A D | LinearAlgebra.cu | 111 if (!iter.can_use_32bit_indexing()) { in unpack_pivots_cuda_kernel()
|
| H A D | UnfoldBackwardKernel.cu | 67 if (!iter.can_use_32bit_indexing()) { in _unfold_backward_internal_kernel()
|
| H A D | SoftMax.cu | 790 …const bool can_use_32bit_indexing = is_32bit_representable(shift) && is_32bit_representable(output… in cunn_SoftMaxBackward() local
792 if (can_use_32bit_indexing) { in cunn_SoftMaxBackward()
805 if (can_use_32bit_indexing) { in cunn_SoftMaxBackward()
811 if (can_use_32bit_indexing) { in cunn_SoftMaxBackward()
|
| H A D | SparseBinaryOpIntersectionKernel.cu | 82 if (!iter.can_use_32bit_indexing()) { in binary_op_intersection_kernel()
|
| H A D | CUDAJitLoops.cuh | 185 TORCH_INTERNAL_ASSERT(iter.can_use_32bit_indexing()); in jitted_gpu_kernel_generic()
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/cuda/ |
| H A D | jiterator.cu | 192 TORCH_INTERNAL_ASSERT(iter.can_use_32bit_indexing()); in jitted_gpu_kernel_dynamic_impl()
296 if (!iter.can_use_32bit_indexing()) { in jitted_gpu_kernel_dynamic()
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/ |
| H A D | TensorIterator.cpp | 1293 bool TensorIteratorBase::can_use_32bit_indexing() const { in can_use_32bit_indexing() function in at::TensorIteratorBase
1682 while (!vec.empty() && !vec.back()->can_use_32bit_indexing()) { in operator ++()
|
| H A D | TensorIterator.h | 503 bool can_use_32bit_indexing() const;
|
| /aosp_15_r20/external/pytorch/torch/_inductor/codegen/ |
| H A D | simd.py | 1169 def can_use_32bit_indexing( member in SIMDScheduling
1231 if SIMDScheduling.can_use_32bit_indexing(total_numel, buffers):
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/cpu/ |
| H A D | UnaryOpsKernel.cpp | 98 if (!it->can_use_32bit_indexing()) { in LogitMKLKernel()
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/mps/ |
| H A D | OperationUtils.mm | 772 …TORCH_CHECK(use_64bit_index || iter.can_use_32bit_indexing(), "Can't be indexed using 32-bit itera…
|
| /aosp_15_r20/external/pytorch/torch/_inductor/ |
| H A D | select_algorithm.py | 644 if not TritonScheduling.can_use_32bit_indexing(numel, buffers):
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/mps/operations/ |
| H A D | Indexing.mm | 100 const bool use_64bit_indexing = !iter.can_use_32bit_indexing();
|