Home
last modified time | relevance | path

Searched refs:grid_z (Results 1 – 25 of 27) sorted by relevance

12

/external/tensorflow/tensorflow/lite/delegates/gpu/common/task/
Dwork_group_picking.cc199 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()
Dgpu_operation.cc465 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
466 return int3(grid_x, grid_y, grid_z); in GetGridSize()
471 const int grid_z = 1; in GetGridSize() local
472 return int3(grid_x, grid_y, grid_z); in GetGridSize()
477 const int grid_z = dst_[0]->Depth(); in GetGridSize() local
478 return int3(grid_x, grid_y, grid_z); in GetGridSize()
483 const int grid_z = 1; in GetGridSize() local
484 return int3(grid_x, grid_y, grid_z); in GetGridSize()
/external/tensorflow/tensorflow/lite/delegates/gpu/common/tasks/
Dresize.cc104 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
105 return int3(grid_x, grid_y, grid_z); in GetGridSize()
203 const int grid_z = dst_[0]->Slices() * dst_[0]->Depth(); in GetGridSize() local
204 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dcumsum.cc95 const int grid_z = slices; in GetGridSize() local
96 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dstrided_slice.cc173 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
174 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dconvolution_transposed_thin.cc153 const int grid_z = 1; in GetGridSize() local
154 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Ddepthwise_conv_3x3_stride_h2.cc199 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
200 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dconv_weights_converter.cc236 const int grid_z = weights_shape.w * weights_shape.h; in GetGridSize() local
237 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dwinograd.cc509 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
510 return int3(grid_x, grid_y, grid_z); in GetGridSize()
727 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
728 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dsplit.cc207 const int grid_z = slices; in GetGridSize() local
208 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dconvolution_transposed_3x3_thin.cc210 const int grid_z = 1; in GetGridSize() local
211 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Ddepthwise_conv_3x3.cc301 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
302 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dreduce.cc504 int grid_z = dst_[0]->Slices(); in GetGridSize() local
508 grid_z *= work_group_size_.z; in GetGridSize()
510 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dconvolution_transposed_4x4.cc363 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
364 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dmean_stddev_normalization.cc328 const int grid_z = work_group_size_.z; in GetGridSize() local
329 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dconvolution_transposed_3x3.cc361 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
362 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Ddepthwise_conv.cc126 const int grid_z = dst_[0]->Slices(); in GetGridSize() local
127 return int3(grid_x, grid_y, grid_z); in GetGridSize()
Dconv_generic.cc1142 int grid_z = DivideRoundUp(dst_slices, block_size.w); in GetGroupsCount() local
1145 DivideRoundUp(grid_z, wg_size.z); in GetGroupsCount()
1154 int grid_z = DivideRoundUp(dst_slices, block_size.w); in GetGroupsCountForLinearWH() local
1157 DivideRoundUp(grid_z, wg_size.y); in GetGroupsCountForLinearWH()
1166 int grid_z = DivideRoundUp(dst_slices, block_size.w); in GetGroupsCountForLinearWHS() local
1168 return DivideRoundUp(grid_x * grid_y * grid_z, wg_size.x); in GetGroupsCountForLinearWHS()
Dconvolution_transposed.cc597 const int grid_z = DivideRoundUp(dst_[0]->Slices(), block_size_.w); in GetGridSize() local
598 return int3(grid_x, grid_y, grid_z); in GetGridSize()
/external/tensorflow/tensorflow/core/profiler/backends/gpu/
Dcupti_collector.h104 uint32 grid_z; member
119 kernel_info.grid_z, " block:", kernel_info.block_x, ",", in ToXStat()
Drocm_tracer.cc468 event.kernel_info.grid_z = data->args.hipModuleLaunchKernel.gridDimZ; in AddKernelEventUponApiExit()
494 event.kernel_info.grid_z = in AddKernelEventUponApiExit()
515 event.kernel_info.grid_z = in AddKernelEventUponApiExit()
537 event.kernel_info.grid_z = data->args.hipLaunchKernel.numBlocks.z; in AddKernelEventUponApiExit()
Drocm_tracer.h73 uint32_t grid_z; member
/external/tensorflow/tensorflow/core/profiler/utils/
Dkernel_stats_utils_test.cc80 kernel_info.grid_z = 1; in TEST()
/external/mesa3d/src/gallium/drivers/llvmpipe/
Dlp_jit.h400 uint32_t grid_z,
Dlp_state_cs.c1277 unsigned grid_z = iter_idx / (job_info->grid_size[0] * job_info->grid_size[1]); in cs_exec_fn() local
1278 …unsigned grid_y = (iter_idx - (grid_z * (job_info->grid_size[0] * job_info->grid_size[1]))) / job_… in cs_exec_fn()
1279 …unsigned grid_x = (iter_idx - (grid_z * (job_info->grid_size[0] * job_info->grid_size[1])) - (grid… in cs_exec_fn()
1283 grid_x, grid_y, grid_z, in cs_exec_fn()

12