/external/tensorflow/tensorflow/core/util/ |
D | work_sharder.cc | 33 int64 cost_per_unit, std::function<void(int64, int64)> work) { in Shard() argument 45 workers->ParallelFor(total, cost_per_unit, work); in Shard() 48 Sharder::Do(total, cost_per_unit, work, in Shard() 55 void Sharder::Do(int64 total, int64 cost_per_unit, const Work& work, in Do() argument 57 cost_per_unit = std::max(int64{1}, cost_per_unit); in Do() 67 total * cost_per_unit / kMinCostPerShard)); in Do()
|
D | work_sharder_test.cc | 30 void RunSharding(int64 num_workers, int64 total, int64 cost_per_unit, in RunSharding() argument 37 Shard(num_workers, threads, total, cost_per_unit, in RunSharding() 50 LOG(INFO) << num_workers << " " << total << " " << cost_per_unit << " " in RunSharding() 68 for (auto cost_per_unit : {0, 1, 11, 102, 1003, 10005, 1000007}) { in TEST() 71 RunSharding(workers, total, cost_per_unit, maxp, &threads); in TEST() 82 const int64 cost_per_unit = 10; in TEST() local 84 Shard(workers, &threads, total_elements, cost_per_unit, in TEST()
|
D | work_sharder.h | 58 int64 cost_per_unit, std::function<void(int64, int64)> work); 95 static void Do(int64 total, int64 cost_per_unit, const Work& work,
|
/external/tensorflow/tensorflow/core/lib/core/ |
D | threadpool.cc | 96 void ParallelFor(int64 total, int64 cost_per_unit, in ParallelFor() 102 total, Eigen::TensorOpCost(0, 0, cost_per_unit), in ParallelFor() 179 void ThreadPool::ParallelFor(int64 total, int64 cost_per_unit, in ParallelFor() argument 181 impl_->ParallelFor(total, cost_per_unit, std::move(fn)); in ParallelFor() 185 int64 total, int64 cost_per_unit, in ParallelForWithWorkerId() argument 187 impl_->ParallelFor(total, cost_per_unit, in ParallelForWithWorkerId()
|
D | threadpool.h | 94 void ParallelFor(int64 total, int64 cost_per_unit, 113 int64 total, int64 cost_per_unit,
|
D | threadpool_test.cc | 239 static void BM_ParallelFor(int iters, int total, int cost_per_unit) { in BM_ParallelFor() argument 248 total, cost_per_unit, in BM_ParallelFor()
|
/external/tensorflow/tensorflow/contrib/nearest_neighbor/kernels/ |
D | hyperplane_lsh_probes.cc | 119 int64 cost_per_unit = 21 * num_hyperplanes_per_table * num_tables; in Compute() local 121 cost_per_unit += in Compute() 125 batch_size, cost_per_unit, [&](int64 start, int64 end) { in Compute()
|
/external/tensorflow/tensorflow/core/kernels/ |
D | concat_lib_cpu.h | 36 int64 cost_per_unit, ElementCopier copier, in ConcatCPUImpl() argument 129 cost_per_unit, work); in ConcatCPUImpl() 138 int64 cost_per_unit, ElementCopier copier, in ConcatSYCLImpl() argument
|
D | reverse_op.cc | 79 const int64 cost_per_unit = input.NumElements() / N; in ReverseRows() local 81 Shard(worker_threads->num_threads, worker_threads->workers, N, cost_per_unit, in ReverseRows()
|
D | batch_matmul_op_impl.h | 211 const int64 cost_per_unit = 218 (batch_size == 1 || cost_per_unit > kMaxCostOuterParallelism)) { 229 cost_per_unit,
|
/external/tensorflow/tensorflow/contrib/coder/kernels/ |
D | pmf_to_cdf_op.cc | 68 const int64 cost_per_unit = static_cast<int64>(50.0 * n * std::log2(n)); in Compute() local 72 pmf.dimension(0), cost_per_unit, in Compute()
|
/external/tensorflow/tensorflow/contrib/factorization/kernels/ |
D | masked_matmul_ops.cc | 173 const int64 cost_per_unit = 20 * a_dim_1; in Compute() local 184 worker_threads.workers->ParallelFor(num_nonzero_elements, cost_per_unit, in Compute()
|