Searched refs:num_threads_y (Results 1 – 3 of 3) sorted by relevance
| /external/tensorflow/tensorflow/compiler/xla/service/llvm_ir/ |
| D | kernel_tiling.cc | 120 absl::Span<const int64> req_block_sizes, int64 num_threads_y, in KernelMappingScheme() argument
126 num_threads_y_(num_threads_y),
|
| D | kernel_tiling.h | 101 int64 num_threads_y, int64 num_threads_x,
|
| /external/tensorflow/tensorflow/compiler/xla/service/gpu/ |
| D | ir_emitter_unnested.cc | 2322 int64 num_threads_y = mapping_scheme->GetNumberOfThreadsForDimensionY(); in EmitFullElementalTile() local
2335 /*step=*/llvm::ConstantInt::get(index_ty, num_threads_y), in EmitFullElementalTile()
2362 int64 num_threads_y = mapping_scheme->GetNumberOfThreadsForDimensionY(); in EmitPartialElementalTile() local
2386 index_ty, num_threads_y - 1)), in EmitPartialElementalTile()
2387 llvm::ConstantInt::get(index_ty, num_threads_y)); in EmitPartialElementalTile()
2390 llvm::ConstantInt::get(index_ty, num_threads_y)); in EmitPartialElementalTile()
2394 /*step=*/llvm::ConstantInt::get(index_ty, num_threads_y), in EmitPartialElementalTile()
3659 int64 num_threads_y = 1; in ComputeMappingSchemeAndReductionKind() local
3723 dims_in_elem, tile_size_y, tile_size_x, req_block_sizes, num_threads_y, in ComputeMappingSchemeAndReductionKind()
|