Home
last modified time | relevance | path

Searched refs:log_num_threads_x (Results 1 – 1 of 1) sorted by relevance

/aosp_15_r20/external/pytorch/aten/src/ATen/native/cuda/
H A DScanUtils.cuh21 integer log_num_threads_x = 0; in get_log_num_threads_x_inner_scan() local
23 while (((integer)1 << log_num_threads_x) < row_size) { in get_log_num_threads_x_inner_scan()
24 ++log_num_threads_x; in get_log_num_threads_x_inner_scan()
32 integer diff = log_num_threads_x - log_num_threads_y; in get_log_num_threads_x_inner_scan()
34 log_num_threads_x = ((integer)9 + diff) / (integer)2; in get_log_num_threads_x_inner_scan()
39 log_num_threads_x = std::min(std::max((integer)4, log_num_threads_x), (integer)9); in get_log_num_threads_x_inner_scan()
40 return log_num_threads_x; in get_log_num_threads_x_inner_scan()
63 … const uint32_t num_threads, const uint32_t log_num_threads_x, in tensor_kernel_scan_innermost_dim_with_indices() argument
69 const uint32_t num_threads_x = 1 << log_num_threads_x; in tensor_kernel_scan_innermost_dim_with_indices()
227 const uint32_t log_num_threads_x = get_log_num_threads_x_inner_scan<uint32_t>(num_rows, row_size); in scan_innermost_dim_with_indices() local
[all …]