Searched refs:work_element_count (Results 1 – 3 of 3) sorted by relevance
/external/tensorflow/tensorflow/core/util/ |
D | cuda_launch_config.h | 125 inline CudaLaunchConfig GetCudaLaunchConfig(int work_element_count, in GetCudaLaunchConfig() argument 127 CHECK_GT(work_element_count, 0); in GetCudaLaunchConfig() 129 const int virtual_thread_count = work_element_count; in GetCudaLaunchConfig() 148 inline CudaLaunchConfig GetCudaLaunchConfig(int work_element_count, in GetCudaLaunchConfig() argument 153 CHECK_GT(work_element_count, 0); in GetCudaLaunchConfig() 164 std::min(block_count, DivUp(work_element_count, thread_per_block)); in GetCudaLaunchConfig() 166 config.virtual_thread_count = work_element_count; in GetCudaLaunchConfig() 178 int work_element_count, const Eigen::GpuDevice& d, DeviceFunc func, in GetCudaLaunchConfigFixedBlockSize() argument 180 CHECK_GT(work_element_count, 0); in GetCudaLaunchConfigFixedBlockSize() 188 DivUp(work_element_count, fixed_block_size)); in GetCudaLaunchConfigFixedBlockSize() [all …]
|
D | cuda_kernel_helper_test.cu.cc | 155 #define TEST_LAUNCH_PARAMETER(work_element_count) \ in TEST_F() argument 161 cfg = GetCudaLaunchConfig(work_element_count, d); \ in TEST_F() 165 EXPECT_EQ(work_element_count, std::accumulate(outbuf, outbuf + bufsize, 0)); \ in TEST_F() 172 cfg = GetCudaLaunchConfig(work_element_count, d, Count1D, 0, 0); \ in TEST_F() 176 EXPECT_EQ(work_element_count, std::accumulate(outbuf, outbuf + bufsize, 0)) in TEST_F()
|
/external/tensorflow/tensorflow/core/kernels/ |
D | stateful_random_ops_gpu.cu.cc | 71 int work_element_count = (output_size + kGroupSize - 1) / kGroupSize; in operator ()() local 72 CudaLaunchConfig cfg = GetCudaLaunchConfig(work_element_count, d, in operator ()()
|