/external/tensorflow/tensorflow/lite/delegates/gpu/common/task/ |
D | gpu_operation.cc | 219 const int grid_y = dst_[0]->Height() * dst_[0]->Depth(); in GetGridSize() local 221 return int3(grid_x, grid_y, grid_z); in GetGridSize() 225 const int grid_y = dst_[0]->Height() * dst_[0]->Depth(); in GetGridSize() local 227 return int3(grid_x, grid_y, grid_z); in GetGridSize() 231 const int grid_y = dst_[0]->Height(); in GetGridSize() local 233 return int3(grid_x, grid_y, grid_z); in GetGridSize() 237 const int grid_y = 1; in GetGridSize() local 239 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
/external/llvm-project/mlir/test/Dialect/GPU/ |
D | outlining.mlir | 26 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %gDimX, %grid_y = %gDimY, 68 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %cst, %grid_y = %cst, 99 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %cst, %grid_y = %cst, 123 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %cst, %grid_y = %cst, 151 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %cst, %grid_y = %cst, 178 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %c1, %grid_y = %c1, 204 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %c1, %grid_y = %c1, 223 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %cst, %grid_y = %cst,
|
D | multiple-all-reduce.mlir | 10 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %c1, %grid_y = %c1, %grid_z = %c1)
|
/external/tensorflow/tensorflow/lite/delegates/gpu/common/tasks/ |
D | resize.cc | 148 const int grid_y = dst_[0]->Height(); in GetGridSize() local 150 return int3(grid_x, grid_y, grid_z); in GetGridSize() 305 const int grid_y = dst_[0]->Height(); in GetGridSize() local 307 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
D | mean_stddev_normalization.cc | 212 const int grid_y = src_[0]->Batch(); in GetGridSize() local 214 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
D | conv_metal.cc | 625 int grid_y = DivideRoundUp(dst_shape.h, block_size.y); in GetGroupsCount() local 628 return DivideRoundUp(grid_x, wg_size.x) * DivideRoundUp(grid_y, wg_size.y) * in GetGroupsCount() 637 int grid_y = DivideRoundUp(dst_shape.h, block_size.y); in GetGroupsCountForLinearWH() local 640 return DivideRoundUp(grid_x * grid_y, wg_size.x) * in GetGroupsCountForLinearWH() 649 int grid_y = DivideRoundUp(dst_shape.h, block_size.y); in GetGroupsCountForLinearWHS() local 652 return DivideRoundUp(grid_x * grid_y * grid_z, wg_size.x); in GetGroupsCountForLinearWHS() 962 const int grid_y = DivideRoundUp(dst_[0]->Height(), params_.block_size.y); in BindArguments() local 964 RETURN_IF_ERROR(args->SetInt("task_size_y", grid_x * grid_y)); in BindArguments() 971 int grid_y = DivideRoundUp(dst_[0]->Height(), params_.block_size.y); in GetGridSize() local 978 return int3(grid_x * grid_y * grid_z, 1, 1); in GetGridSize() [all …]
|
D | split.cc | 146 const int grid_y = height * depth; in GetGridSize() local 148 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
D | strided_slice.cc | 171 const int grid_y = dst_[0]->Height(); in GetGridSize() local 173 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
D | conv_weights_converter.cc | 182 const int grid_y = src_[0]->Slices(); in GetGridSize() local 184 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
D | depthwise_conv_3x3_stride_h2.cc | 192 const int grid_y = DivideRoundUp(dst_[0]->Height(), 2); in GetGridSize() local 194 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
D | convolution_transposed_thin.cc | 152 const int grid_y = src_[0]->Height(); in GetGridSize() local 154 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
D | winograd.cc | 462 const int grid_y = 6; in GetGridSize() local 464 return int3(grid_x, grid_y, grid_z); in GetGridSize() 676 const int grid_y = 4; in GetGridSize() local 678 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
D | convolution_transposed_3x3_thin.cc | 194 const int grid_y = src_[0]->Height(); in GetGridSize() local 196 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
D | depthwise_conv_3x3.cc | 286 const int grid_y = DivideRoundUp(dst_[0]->Height(), 2); in GetGridSize() local 288 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
D | reduce.cc | 453 int grid_y = dst_[0]->Height() * dst_[0]->Depth(); in GetGridSize() local 457 grid_y *= work_group_size_.y; in GetGridSize() 460 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
/external/llvm-project/mlir/test/mlir-cuda-runner/ |
D | two-modules.mlir | 12 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %one, %grid_y = %one, %grid_z = %one) 18 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %one, %grid_y = %one, %grid_z = %one)
|
D | gpu-to-cubin.mlir | 7 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %cst, %grid_y = %cst, %grid_z = %cst)
|
D | all-reduce-region.mlir | 12 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %one, %grid_y = %one, %grid_z = %one)
|
D | shuffle.mlir | 12 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %one, %grid_y = %one, %grid_z = %one)
|
D | all-reduce-op.mlir | 15 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %c1, %grid_y = %c1, %grid_z = %c1)
|
/external/llvm-project/mlir/test/mlir-rocm-runner/ |
D | two-modules.mlir | 13 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %c1, %grid_y = %c1, %grid_z = %c1) 19 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %c1, %grid_y = %c1, %grid_z = %c1)
|
D | vector-transferops.mlir | 5 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %cst, %grid_y = %cst, %grid_z = %cst) 27 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %cst, %grid_y = %cst, %grid_z = %cst)
|
D | gpu-to-hsaco.mlir | 7 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %c1, %grid_y = %c1, %grid_z = %c1)
|
D | vecadd.mlir | 7 gpu.launch blocks(%bx, %by, %bz) in (%grid_x = %c1, %grid_y = %c1, %grid_z = %c1)
|
/external/tensorflow/tensorflow/core/profiler/internal/gpu/ |
D | cupti_collector.h | 93 uint32 grid_y; member 104 " grid:", kernel_info.grid_x, ",", kernel_info.grid_y, ",", in ToXStat()
|