| /external/tensorflow/tensorflow/lite/delegates/gpu/common/task/ |
| D | gpu_operation.cc | 220 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
221 return int3(grid_x, grid_y, grid_z); in GetGridSize()
226 const int grid_z = 1; in GetGridSize() local
227 return int3(grid_x, grid_y, grid_z); in GetGridSize()
232 const int grid_z = dst_[0]->Depth(); in GetGridSize() local
233 return int3(grid_x, grid_y, grid_z); in GetGridSize()
238 const int grid_z = 1; in GetGridSize() local
239 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | work_group_picking.cc | 199 int grid_z = GetBiggestDividerWithPriority(grid.z, 4); in GetWorkGroupXY128ConvLinear() local
201 return int3(128, 1, grid_z); in GetWorkGroupXY128ConvLinear()
203 int grid_x = GetMaxSizeWithMinPenalty(grid.x, 512 / grid_z); in GetWorkGroupXY128ConvLinear()
204 return {grid_x, 1, grid_z}; in GetWorkGroupXY128ConvLinear()
208 int grid_z = GetBiggestDividerWithPriority(grid.z, 4); in GetWorkGroupXY128Conv() local
210 return int3(16, 8, grid_z); in GetWorkGroupXY128Conv()
212 int2 grid_xy = GetMaxSizeWithMinPenalty(int2(grid.x, grid.y), 512 / grid_z); in GetWorkGroupXY128Conv()
213 return int3(grid_xy.x, grid_xy.y, grid_z); in GetWorkGroupXY128Conv()
|
| /external/llvm-project/mlir/test/Dialect/GPU/ |
| D | outlining.mlir | 27 %grid_z = %gDimZ)
69 %grid_z = %cst)
100 %grid_z = %cst)
124 %grid_z = %cst)
152 %grid_z = %cst)
179 %grid_z = %c1)
205 %grid_z = %c1)
224 %grid_z = %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 | 149 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
150 return int3(grid_x, grid_y, grid_z); in GetGridSize()
306 const int grid_z = dst_[0]->Slices() * dst_[0]->Depth(); in GetGridSize() local
307 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | mean_stddev_normalization.cc | 213 const int grid_z = 1; in GetGridSize() local
214 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | split.cc | 147 const int grid_z = slices; in GetGridSize() local
148 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | strided_slice.cc | 172 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
173 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | conv_weights_converter.cc | 183 const int grid_z = src_[0]->Width() * src_[0]->Height(); in GetGridSize() local
184 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | conv_metal.cc | 626 int grid_z = DivideRoundUp(dst_slices, block_size.z); in GetGroupsCount() local
629 DivideRoundUp(grid_z, wg_size.z); in GetGroupsCount()
638 int grid_z = DivideRoundUp(dst_slices, block_size.z); in GetGroupsCountForLinearWH() local
641 DivideRoundUp(grid_z, wg_size.y); in GetGroupsCountForLinearWH()
650 int grid_z = DivideRoundUp(dst_slices, block_size.z); in GetGroupsCountForLinearWHS() local
652 return DivideRoundUp(grid_x * grid_y * grid_z, wg_size.x); in GetGroupsCountForLinearWHS()
972 int grid_z = DivideRoundUp(dst_[0]->Slices(), params_.block_size.z); in GetGridSize() local
978 return int3(grid_x * grid_y * grid_z, 1, 1); in GetGridSize()
980 return int3(grid_x * grid_y, grid_z, 1); in GetGridSize()
982 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | depthwise_conv_3x3_stride_h2.cc | 193 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
194 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | convolution_transposed_thin.cc | 153 const int grid_z = 1; in GetGridSize() local
154 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | winograd.cc | 463 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
464 return int3(grid_x, grid_y, grid_z); in GetGridSize()
677 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
678 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | convolution_transposed_3x3_thin.cc | 195 const int grid_z = 1; in GetGridSize() local
196 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | depthwise_conv_3x3.cc | 287 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
288 return int3(grid_x, grid_y, grid_z); in GetGridSize()
|
| D | reduce.cc | 454 int grid_z = dst_[0]->Slices(); in GetGridSize() local
458 grid_z *= work_group_size_.z; 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)
|
| /external/tensorflow/tensorflow/core/profiler/internal/gpu/ |
| D | cupti_collector.h | 95 uint32 grid_z; member
105 kernel_info.grid_z, " block:", kernel_info.block_x, ",", in ToXStat()
|