MCPcopy Create free account

hub / github.com/ROCm/rocPRIM / functions

Functions2,472 in github.com/ROCm/rocPRIM

↓ 1 callersFunctionimport_module_from_file
Import a module from a file path.
tuning/run_tuning.py:46
↓ 1 callersFunctioninclusive_scan_by_key
rocprim/include/rocprim/device/device_scan_by_key.hpp:473
↓ 1 callersFunctioninit_with_run_lengths
rocprim/include/rocprim/block/block_run_length_decode.hpp:268
↓ 1 callersMethodis
benchmark/cmdparser.hpp:75
↓ 1 callersFunctionis_warp_size_supported
benchmark/benchmark_utils.hpp:449
↓ 1 callersMethodlaunch
rocprim/include/rocprim/device/device_histogram.hpp:138
↓ 1 callersMethodlaunchGraph
test/rocprim/test_utils_hipgraphs.hpp:52
↓ 1 callersMethodload
rocprim/include/rocprim/warp/warp_load.hpp:166
↓ 1 callersMethodload_input_values
rocprim/include/rocprim/device/detail/device_run_length_encode.hpp:62
↓ 1 callersFunctionload_json_files
Load and process JSON files from both simulated and actual output directories. Returns best times for each kernel configuration.
tuning/utils/compare_simulated_strategies.py:110
↓ 1 callersMethodload_keys_values
rocprim/include/rocprim/device/detail/device_reduce_by_key.hpp:88
↓ 1 callersFunctionload_selected_count
rocprim/include/rocprim/device/detail/device_partition.hpp:750
↓ 1 callersFunctionmain
()
rtest.py:338
↓ 1 callersFunctionmain
()
rmake.py:245
↓ 1 callersFunctionmain
()
tuning/configs_header_generation.py:1091
↓ 1 callersFunctionmain
()
tuning/run_tuning.py:157
↓ 1 callersFunctionmain
()
tuning/utils/compare_simulated_strategies.py:237
↓ 1 callersFunctionmain
()
scripts/apply_config_improvements/apply_config_improvements.py:732
↓ 1 callersFunctionmain
()
scripts/autotune/create_optimization.py:876
↓ 1 callersFunctionmake_cmd
()
rmake.py:207
↓ 1 callersFunctionmemcpy_and_sync
* \brief Copy data from src to dest with stream ordering and synchronization * * Equivalent to `hipStreamMemcpyAsync(...,stream)` followed by `hipSt
rocprim/include/rocprim/detail/various.hpp:303
↓ 1 callersMethodmerge
rocprim/include/rocprim/device/detail/device_merge.hpp:272
↓ 1 callersFunctionmerge_impl
rocprim/include/rocprim/device/device_merge.hpp:81
↓ 1 callersMethodname
benchmark/benchmark_device_search_n.parallel.hpp:98
↓ 1 callersMethodname
benchmark/benchmark_device_binary_search.parallel.hpp:55
↓ 1 callersFunctionoperator<<
test/rocprim/test_utils_assertions.hpp:394
↓ 1 callersFunctionos_detect
()
rtest.py:69
↓ 1 callersFunctionos_detect
()
rmake.py:50
↓ 1 callersMethodoutput_specialization_info
This function doesn't use a JSON library, since it just appends text onto the end of a JSON file. Seeking and writing to the end of a huge file is ver
benchmark/benchmark_utils.hpp:1540
↓ 1 callersFunctionpad_string
Inserts spaces at beginning of string if string shorter than specified length.
benchmark/benchmark_utils.hpp:571
↓ 1 callersFunctionparse_args
Parse command-line arguments
rtest.py:28
↓ 1 callersFunctionparse_args
Parse command-line arguments
rmake.py:18
↓ 1 callersFunctionparse_arguments
()
tuning/utils/compare_simulated_strategies.py:64
↓ 1 callersMethodparse_header
Parse the header content and extract configurations. Args: header_content: Content of the header file. Returns:
tuning/utils/config_parser.py:109
↓ 1 callersMethodpartition
rocprim/include/rocprim/device/detail/device_partition.hpp:970
↓ 1 callersFunctionpartition_three_way
rocprim/include/rocprim/device/device_partition.hpp:1069
↓ 1 callersFunctionpartition_two_way
rocprim/include/rocprim/device/device_partition.hpp:483
↓ 1 callersMethodprepare
benchmark/benchmark_device_merge_inplace.cpp:113
↓ 1 callersFunctionprint_results
()
scripts/apply_config_improvements/apply_config_improvements.py:54
↓ 1 callersMethodprint_value
benchmark/cmdparser.hpp:109
↓ 1 callersMethodqueue_fn
benchmark/benchmark_utils.hpp:1956
↓ 1 callersFunctionradix_rank_algorithm_name
benchmark/benchmark_device_radix_sort_onesweep.parallel.hpp:58
↓ 1 callersMethodrecord_amdsmi_stats
benchmark/benchmark_utils.hpp:1512
↓ 1 callersFunctionreduce_by_key
rocprim/include/rocprim/device/device_reduce_by_key.hpp:449
↓ 1 callersFunctionreduce_impl
rocprim/include/rocprim/device/device_reduce.hpp:99
↓ 1 callersMethodresetGraphHelper
test/rocprim/test_utils_hipgraphs.hpp:80
↓ 1 callersMethodresize_async
common/utils_device_ptr.hpp:262
↓ 1 callersMethodresolve
benchmark/benchmark_device_search_n.parallel.hpp:102
↓ 1 callersMethodrun
benchmark/benchmark_device_segmented_reduce.parallel.hpp:183
↓ 1 callersMethodrun
benchmark/benchmark_device_segmented_radix_sort_keys.parallel.hpp:201
↓ 1 callersMethodrun
benchmark/benchmark_block_exchange.cpp:60
↓ 1 callersMethodrun
benchmark/benchmark_device_segmented_radix_sort_pairs.parallel.hpp:215
↓ 1 callersMethodrun_and_exit_if_error
benchmark/cmdparser.hpp:322
↓ 1 callersFunctionrun_tests
()
rtest.py:316
↓ 1 callersFunctionrun_tuning
Run tuning for specified algorithms.
tuning/run_tuning.py:80
↓ 1 callersMethodsave_batch_times
benchmark/benchmark_utils.hpp:1521
↓ 1 callersMethodscan
rocprim/include/rocprim/device/detail/device_run_length_encode.hpp:350
↓ 1 callersMethodscatter
rocprim/include/rocprim/device/detail/device_run_length_encode.hpp:248
↓ 1 callersFunctionselect_previous_values_iterator
rocprim/include/rocprim/device/detail/device_adjacent_difference.hpp:153
↓ 1 callersFunctionserial_merge
rocprim/include/rocprim/detail/merge_path.hpp:102
↓ 1 callersFunctionserial_merge
test/rocprim/internal/test_internal_merge_path.cpp:23
↓ 1 callersMethodserialize_context
benchmark/benchmark_utils.hpp:1072
↓ 1 callersMethodserialize_stats
benchmark/benchmark_utils.hpp:1048
↓ 1 callersMethodset_partial
rocprim/include/rocprim/device/detail/lookback_scan_state.hpp:338
↓ 1 callersMethodsort_and_scatter
rocprim/include/rocprim/device/detail/device_radix_sort.hpp:381
↓ 1 callersFunctionsort_impl
rocprim/include/rocprim/block/block_radix_sort.hpp:1143
↓ 1 callersMethodstopped
(self)
rtest.py:132
↓ 1 callersMethodstore
rocprim/include/rocprim/device/detail/device_partition.hpp:791
↓ 1 callersMethodstore
rocprim/include/rocprim/warp/warp_store.hpp:170
↓ 1 callersFunctionstore_selected_count
rocprim/include/rocprim/device/detail/device_partition.hpp:730
↓ 1 callersFunctiontestHistogramEvenIncorrectInput
test/rocprim/test_device_histogram.cpp:177
↓ 1 callersFunctiontestHistogramRangeIncorrectInput
test/rocprim/test_device_histogram.cpp:391
↓ 1 callersFunctiontestLargeIndices
test/rocprim/test_device_transform.cpp:448
↓ 1 callersFunctiontestLargeIndices
test/rocprim/test_device_merge_sort.cpp:415
↓ 1 callersFunctiontestLargeIndicesExclusiveScan
test/rocprim/test_device_scan.cpp:1305
↓ 1 callersFunctiontestLargeIndicesExclusiveScanByKey
test/rocprim/test_device_scan_by_key.cpp:970
↓ 1 callersFunctiontestLargeIndicesInclusiveScan
test/rocprim/test_device_scan.cpp:1187
↓ 1 callersFunctiontestLargeIndicesInclusiveScanByKey
test/rocprim/test_device_scan_by_key.cpp:928
↓ 1 callersFunctiontestManualConstruction
test/hipgraph/test_hipgraph_basic.cpp:108
↓ 1 callersFunctiontestStreamCapture
test/hipgraph/test_hipgraph_basic.cpp:53
↓ 1 callersFunctiontestStreamCaptureWithAtomics
test/hipgraph/test_hipgraph_basic.cpp:168
↓ 1 callersFunctiontestUniqueGuardedOperator
test/rocprim/test_device_select.cpp:683
↓ 1 callersFunctiontransform_nary
test/rocprim/test_device_transform.cpp:312
↓ 1 callersFunctiontranslate_settings_to_cpp_metaprogramming
Translates a list of named fallback configuration entries to C++ metaprogramming idioms.
tuning/configs_header_generation.py:118
↓ 1 callersFunctiontranslate_settings_to_cpp_metaprogramming
Translates a list of named fallback configuration entries to C++ metaprogramming idioms.
scripts/autotune/create_optimization.py:104
↓ 1 callersFunctiontune_alg
The core tuning procedure. This tunes a single algorithm for multiple types.
scripts/autotune-search/main.py:155
↓ 1 callersMethodunpack
rocprim/include/rocprim/detail/various.hpp:580
↓ 1 callersFunctionunrolled_copy_impl
rocprim/include/rocprim/thread/thread_copy.hpp:194
↓ 1 callersFunctionunrolled_tuple_copy
rocprim/include/rocprim/thread/thread_copy.hpp:139
↓ 1 callersFunctionunrolled_tuple_copy_impl
rocprim/include/rocprim/thread/thread_copy.hpp:127
↓ 1 callersFunctionupper_bound_n
rocprim/include/rocprim/device/detail/device_binary_search.hpp:62
↓ 1 callersMethodusage
benchmark/cmdparser.hpp:458
↓ 1 callersMethodvalue
rocprim/include/rocprim/device/detail/device_nth_element.hpp:83
↓ 1 callersFunctionvram_detect
()
rtest.py:50
↓ 1 callersFunctionwarp_any
rocprim/include/rocprim/intrinsics/warp.hpp:74
↓ 1 callersFunctionwarp_sort_config_name
benchmark/benchmark_device_segmented_radix_sort_keys.parallel.hpp:52
↓ 1 callersFunctionwarp_sort_config_name
benchmark/benchmark_device_segmented_radix_sort_pairs.parallel.hpp:52
↓ 1 callersFunctionworking_directory
Context manager for changing the current working directory.
tuning/run_tuning.py:36
↓ 1 callersMethodwrite_configs_to_files
(self, base_dir: str)
tuning/configs_header_generation.py:1080
↓ 1 callersMethodwrite_configs_to_files
For each algorithm, creates a file containing configurations and places these in base_dir.
scripts/autotune/create_optimization.py:863
← previousnext →501–600 of 2,472, ranked by callers