Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/ROCm/rocPRIM
/ functions
Functions
2,472 in github.com/ROCm/rocPRIM
⨍
Functions
2,472
◇
Types & classes
7,856
↓ 2 callers
Function
get_block_scan_algorithm_name
benchmark/benchmark_utils.hpp:945
↓ 2 callers
Method
get_d_input
benchmark/benchmark_device_histogram.parallel.hpp:214
↓ 2 callers
Function
get_device_arch
rocprim/include/rocprim/device/config_types.hpp:547
↓ 2 callers
Method
get_iterator
test/rocprim/test_device_merge_inplace.cpp:150
↓ 2 callers
Function
get_lds_banks_no
Returns the number of LDS (local data share) banks.
rocprim/include/rocprim/detail/various.hpp:153
↓ 2 callers
Method
get_max_size
test/rocprim/test_device_merge_inplace.cpp:156
↓ 2 callers
Function
get_min_lds_size
\brief Returns the minimum LDS size in bytes available on this device architecture.
rocprim/include/rocprim/detail/various.hpp:161
↓ 2 callers
Method
get_parser
( default_bytes: Optional[int] = None, default_iterations: int = 32, default_max_feval
tuning/utils/utils.py:42
↓ 2 callers
Function
get_sizes
test/hip/test_hip_async_copy.cpp:72
↓ 2 callers
Function
get_target_arch_from_name
rocprim/include/rocprim/device/config_types.hpp:234
↓ 2 callers
Method
get_total_work_size
rocprim/include/rocprim/device/device_merge_inplace.hpp:290
↓ 2 callers
Function
host_exclusive_scan
test/rocprim/test_utils.hpp:331
↓ 2 callers
Function
host_exclusive_scan_by_key_impl
test/rocprim/test_utils.hpp:369
↓ 2 callers
Function
host_exclusive_scan_impl
test/rocprim/test_utils.hpp:312
↓ 2 callers
Function
host_inclusive_scan_by_key_impl
test/rocprim/test_utils.hpp:464
↓ 2 callers
Function
host_inclusive_scan_impl
test/rocprim/test_utils.hpp:262
↓ 2 callers
Function
host_reduce
test/rocprim/test_utils.hpp:169
↓ 2 callers
Method
inclusive_scan
rocprim/include/rocprim/block/detail/block_scan_reduce_then_scan.hpp:80
↓ 2 callers
Method
inclusive_scan_impl
rocprim/include/rocprim/block/detail/block_scan_warp_scan.hpp:527
↓ 2 callers
Method
inclusive_scan_impl
rocprim/include/rocprim/block/detail/block_scan_reduce_then_scan.hpp:514
↓ 2 callers
Function
init_histogram
rocprim/include/rocprim/block/block_histogram.hpp:159
↓ 2 callers
Method
is_build_in
\brief Check if the type is a `build_in` type, this function is different from `is_fundamental`, because, by implementing traits, downstream code can
rocprim/include/rocprim/type_traits.hpp:1155
↓ 2 callers
Function
is_matching_file
Check if the JSON file contains both the algorithm name and architecture name.
tuning/utils/compare_simulated_strategies.py:77
↓ 2 callers
Method
is_valid
rocprim/include/rocprim/device/device_merge_inplace.hpp:109
↓ 2 callers
Method
leak
\brief Clean every thing on this instance, which could lead to memory leak. Should call `get()` and free the raw pointer manually
common/utils_device_ptr.hpp:226
↓ 2 callers
Method
left_size
rocprim/include/rocprim/device/device_merge_inplace.hpp:127
↓ 2 callers
Function
lookback_reduce_forward
rocprim/include/rocprim/device/detail/lookback_scan_state.hpp:155
↓ 2 callers
Function
lookback_reduce_forward_init
rocprim/include/rocprim/device/detail/lookback_scan_state.hpp:125
↓ 2 callers
Function
lower_bound_n
rocprim/include/rocprim/device/detail/device_binary_search.hpp:38
↓ 2 callers
Function
make_mask_iterator
rocprim/include/rocprim/iterator/predicate_iterator.hpp:288
↓ 2 callers
Function
match_any
\overload \brief Group active lanes having the same bits of \p label Threads that have the same least significant \p label_bits bits are grouped into
rocprim/include/rocprim/intrinsics/warp.hpp:105
↓ 2 callers
Function
memory_fence_device
rocprim/include/rocprim/intrinsics/thread.hpp:329
↓ 2 callers
Function
merge_jsons
Merges benchmark JSONs. This is used to collect the singular results generated from the various benchmark runs.
scripts/autotune-search/main.py:113
↓ 2 callers
Function
merge_sort
rocprim/include/rocprim/device/device_merge_sort.hpp:937
↓ 2 callers
Method
operator->
rocprim/include/rocprim/iterator/zip_iterator.hpp:198
↓ 2 callers
Method
permute
rocprim/include/rocprim/device/detail/device_merge_sort.hpp:155
↓ 2 callers
Method
process_tile
rocprim/include/rocprim/device/detail/device_reduce_by_key.hpp:333
↓ 2 callers
Method
queue_instance
benchmark/benchmark_utils.hpp:1964
↓ 2 callers
Method
radix_key_codec
rocprim/include/rocprim/type_traits.hpp:1235
↓ 2 callers
Function
rank_keys
rocprim/include/rocprim/block/block_radix_rank.hpp:227
↓ 2 callers
Method
rank_keys_desc
rocprim/include/rocprim/block/detail/block_radix_rank_basic.hpp:266
↓ 2 callers
Function
read_configs
(dir_path: Path)
scripts/apply_config_improvements/apply_config_improvements.py:525
↓ 2 callers
Function
read_data
(dir_path: Path, selectors: Dict[str, List[str]])
scripts/apply_config_improvements/apply_config_improvements.py:590
↓ 2 callers
Function
reduce
rocprim/include/rocprim/device/device_reduce.hpp:418
↓ 2 callers
Method
right_size
rocprim/include/rocprim/device/device_merge_inplace.hpp:133
↓ 2 callers
Method
run
benchmark/benchmark_device_merge.parallel.hpp:240
↓ 2 callers
Method
run
benchmark/benchmark_warp_scan.cpp:105
↓ 2 callers
Method
run
benchmark/benchmark_device_radix_sort.hpp:205
↓ 2 callers
Method
run
benchmark/benchmark_block_discontinuity.cpp:61
↓ 2 callers
Method
run
benchmark/benchmark_device_merge_sort_block_sort.parallel.hpp:190
↓ 2 callers
Method
run
benchmark/benchmark_block_reduce.cpp:56
↓ 2 callers
Method
run
benchmark/benchmark_device_radix_sort_block_sort.parallel.hpp:181
↓ 2 callers
Method
run
benchmark/benchmark_block_scan.cpp:57
↓ 2 callers
Method
run
benchmark/benchmark_device_merge_sort_block_merge.parallel.hpp:304
↓ 2 callers
Method
run
benchmark/benchmark_utils.hpp:1709
↓ 2 callers
Method
run
benchmark/benchmark_block_histogram.cpp:56
↓ 2 callers
Method
run
benchmark/benchmark_device_radix_sort_onesweep.parallel.hpp:259
↓ 2 callers
Method
run
benchmark/benchmark_device_merge_sort.hpp:219
↓ 2 callers
Function
scan
rocprim/include/rocprim/warp/detail/warp_scan_dpp.hpp:251
↓ 2 callers
Method
scatter
rocprim/include/rocprim/device/detail/device_reduce_by_key.hpp:189
↓ 2 callers
Method
scatter_to_striped
rocprim/include/rocprim/block/block_exchange.hpp:972
↓ 2 callers
Method
scatter_to_warp_striped
rocprim/include/rocprim/block/block_exchange.hpp:979
↓ 2 callers
Function
segmented_reduce
rocprim/include/rocprim/device/device_segmented_reduce.hpp:218
↓ 2 callers
Method
set
rocprim/include/rocprim/device/detail/lookback_scan_state.hpp:491
↓ 2 callers
Method
set
rocprim/include/rocprim/device/detail/lookback_scan_state.hpp:823
↓ 2 callers
Method
set_storage
\brief Assigns the final storage for this partition. `storage` is assumed to have the required alignment and size as described by the layout returned
rocprim/include/rocprim/detail/temp_storage.hpp:231
↓ 2 callers
Function
sort_check_compare
test/rocprim/test_utils_sort_checker.hpp:41
↓ 2 callers
Method
sort_desc
rocprim/include/rocprim/block/block_radix_sort.hpp:1383
↓ 2 callers
Method
sort_desc_to_striped
rocprim/include/rocprim/block/block_radix_sort.hpp:1397
↓ 2 callers
Method
sort_desc_warp_striped_to_striped
rocprim/include/rocprim/block/block_radix_sort.hpp:1412
↓ 2 callers
Method
sort_key
benchmark/benchmark_utils.hpp:1918
↓ 2 callers
Method
sort_to_striped
rocprim/include/rocprim/block/block_radix_sort.hpp:1390
↓ 2 callers
Method
sort_warp_striped_to_striped
rocprim/include/rocprim/block/block_radix_sort.hpp:1404
↓ 2 callers
Method
status
rocprim/include/rocprim/device/detail/device_nth_element.hpp:89
↓ 2 callers
Method
status
rocprim/include/rocprim/device/detail/device_radix_sort.hpp:1030
↓ 2 callers
Method
stop
(self)
rtest.py:129
↓ 2 callers
Method
store
rocprim/include/rocprim/device/detail/device_radix_sort.hpp:1043
↓ 2 callers
Method
store_async
common/utils_device_ptr.hpp:404
↓ 2 callers
Function
strip_ansi
(s: str)
scripts/apply_config_improvements/apply_config_improvements.py:73
↓ 2 callers
Function
stripe_vector
test/rocprim/test_warp_store.cpp:216
↓ 2 callers
Function
stripe_vector
test/rocprim/test_warp_load.cpp:228
↓ 2 callers
Method
striped_to_blocked_shuffle
rocprim/include/rocprim/warp/warp_exchange.hpp:1065
↓ 2 callers
Function
swap
rocprim/include/rocprim/functional.hpp:70
↓ 2 callers
Function
thread_scan_exclusive
rocprim/include/rocprim/thread/thread_scan.hpp:81
↓ 2 callers
Function
thread_scan_inclusive
rocprim/include/rocprim/thread/thread_scan.hpp:135
↓ 2 callers
Function
thread_store
rocprim/include/rocprim/thread/thread_store.hpp:140
↓ 2 callers
Method
tune_all
Tune for all key type and value type combinations
tuning/tuning_scripts/tuning_device_merge.py:244
↓ 2 callers
Method
tune_type
Performs auto-tuning for a specific key type and optional value type combination.
tuning/tuning_scripts/base_tuner.py:232
↓ 2 callers
Function
tuple
\brief Implicitly-defined copy constructor.
rocprim/include/rocprim/types/tuple.hpp:570
↓ 2 callers
Function
upper_bound
rocprim/include/rocprim/device/detail/device_histogram.hpp:181
↓ 2 callers
Function
version
\brief The top level rocPRIM namespace.
rocprim/include/rocprim/rocprim.hpp:88
↓ 2 callers
Function
warp_readlane
rocprim/include/rocprim/intrinsics/warp_shuffle.hpp:330
↓ 2 callers
Function
warp_shuffle_xor
rocprim/include/rocprim/intrinsics/warp_shuffle.hpp:216
↓ 1 callers
Function
Transpose
rocprim/include/rocprim/warp/warp_exchange.hpp:151
↓ 1 callers
Method
__add_benchmark_to_algorithm
Adds a single_benchmark execution of a given Algorithm on a given architecture, to the Algorithm object In case the Algorithm object
scripts/autotune/create_optimization.py:831
↓ 1 callers
Method
__get_single_benchmark
Enriches the benchmark the data in single_benchmark with the information stored in the actual name of the particular benchmark run T
scripts/autotune/create_optimization.py:817
↓ 1 callers
Method
__get_target_architecture_from_context
Uses the benchmark run context embedded into the benchmark output json to retrieve the targeted architecture
scripts/autotune/create_optimization.py:806
↓ 1 callers
Method
__load_fallback_entries
Reads in fallback json file to list of dictionaries
tuning/configs_header_generation.py:989
↓ 1 callers
Method
__load_fallback_entries
Reads in fallback json file to list of dictionaries
scripts/autotune/create_optimization.py:783
↓ 1 callers
Method
_allocate_device_memory
DeviceArray wrapper must be used to properly manage GPU memory allocation/deallocation and configure array properties. See HIP Python
tuning/tuning_scripts/base_tuner.py:178
← previous
next →
301–400 of 2,472, ranked by callers