|
template<typename KeyT , typename ValueT , typename NumItemsT > |
DeviceRadixSort & | SortPairs (const KeyT *d_keys_in, KeyT *d_keys_out, const ValueT *d_values_in, ValueT *d_values_out, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename ValueT , typename NumItemsT > |
DeviceRadixSort & | SortPairs (DeviceVector< std::byte > &external_buffer, cub::DoubleBuffer< KeyT > &d_keys, cub::DoubleBuffer< ValueT > &d_values, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename ValueT , typename NumItemsT > |
DeviceRadixSort & | SortPairsDescending (const KeyT *d_keys_in, KeyT *d_keys_out, const ValueT *d_values_in, ValueT *d_values_out, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename ValueT , typename NumItemsT > |
DeviceRadixSort & | SortPairsDescending (cub::DoubleBuffer< KeyT > &d_keys, cub::DoubleBuffer< ValueT > &d_values, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename NumItemsT > |
DeviceRadixSort & | SortKeys (const KeyT *d_keys_in, KeyT *d_keys_out, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename NumItemsT > |
DeviceRadixSort & | SortKeys (cub::DoubleBuffer< KeyT > &d_keys, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename NumItemsT > |
DeviceRadixSort & | SortKeysDescending (const KeyT *d_keys_in, KeyT *d_keys_out, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename NumItemsT > |
DeviceRadixSort & | SortKeysDescending (cub::DoubleBuffer< KeyT > &d_keys, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename ValueT , typename NumItemsT > |
DeviceRadixSort & | SortPairs (void *d_temp_storage, size_t &temp_storage_bytes, const KeyT *d_keys_in, KeyT *d_keys_out, const ValueT *d_values_in, ValueT *d_values_out, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename ValueT , typename NumItemsT > |
DeviceRadixSort & | SortPairs (void *d_temp_storage, size_t &temp_storage_bytes, cub::DoubleBuffer< KeyT > &d_keys, cub::DoubleBuffer< ValueT > &d_values, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename ValueT , typename NumItemsT > |
DeviceRadixSort & | SortPairsDescending (void *d_temp_storage, size_t &temp_storage_bytes, const KeyT *d_keys_in, KeyT *d_keys_out, const ValueT *d_values_in, ValueT *d_values_out, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename ValueT , typename NumItemsT > |
DeviceRadixSort & | SortPairsDescending (void *d_temp_storage, size_t &temp_storage_bytes, cub::DoubleBuffer< KeyT > &d_keys, cub::DoubleBuffer< ValueT > &d_values, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename NumItemsT > |
DeviceRadixSort & | SortKeys (void *d_temp_storage, size_t &temp_storage_bytes, const KeyT *d_keys_in, KeyT *d_keys_out, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename NumItemsT > |
DeviceRadixSort & | SortKeys (void *d_temp_storage, size_t &temp_storage_bytes, cub::DoubleBuffer< KeyT > &d_keys, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename NumItemsT > |
DeviceRadixSort & | SortKeysDescending (void *d_temp_storage, size_t &temp_storage_bytes, const KeyT *d_keys_in, KeyT *d_keys_out, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
template<typename KeyT , typename NumItemsT > |
DeviceRadixSort & | SortKeysDescending (void *d_temp_storage, size_t &temp_storage_bytes, cub::DoubleBuffer< KeyT > &d_keys, NumItemsT num_items, int begin_bit=0, int end_bit=sizeof(KeyT) *8) |
|
| CubWrapper (Stream &stream=Stream::Default()) |
|
void | kernel_name (std::string_view)=delete |
|
MUDA_GENERIC | LaunchBase (::cudaStream_t stream) MUDA_NOEXCEPT |
|
T & | push_range (const std::string &name) |
|
T & | pop_range () |
|
T & | kernel_name (std::string_view name) |
|
T & | file_line (std::string_view file, int line) |
|
T & | record (cudaEvent_t e, int flag=cudaEventRecordDefault) |
|
T & | record (ComputeGraphVar< cudaEvent_t > &e, const std::vector< ComputeGraphVarBase * > &vars) |
|
template<typename... ViewT> |
T & | record (ComputeGraphVar< cudaEvent_t > &e, ComputeGraphVar< ViewT > &... vars) |
|
T & | when (cudaEvent_t e, int flag=cudaEventWaitDefault) |
|
T & | wait (cudaEvent_t e, int flag=cudaEventWaitDefault) |
|
T & | wait (const ComputeGraphVar< cudaEvent_t > &e, const std::vector< ComputeGraphVarBase * > &vars) |
|
template<typename... ViewT> |
T & | wait (const ComputeGraphVar< cudaEvent_t > &e, ComputeGraphVar< ViewT > &... vars) |
|
T & | wait () |
|
T & | callback (const std::function< void(::cudaStream_t, ::cudaError)> &callback) |
|
template<typename Next > |
Next | next (Next n) |
|
template<typename Next , typename... Args> |
Next | next (Args &&... args) |
|
template<typename T > |
MUDA_GENERIC | LaunchBase (cudaStream_t stream) MUDA_NOEXCEPT |
|
MUDA_GENERIC | LaunchCore (::cudaStream_t stream) MUDA_NOEXCEPT |
|
void | init_stream (::cudaStream_t s) |
|
void | push_range (const std::string &name) |
|
void | pop_range () |
|
void | record (cudaEvent_t e, int flag=cudaEventRecordDefault) |
|
void | record (ComputeGraphVar< cudaEvent_t > &e, const std::vector< ComputeGraphVarBase * > &vars) |
|
template<typename... ViewT> |
void | record (ComputeGraphVar< cudaEvent_t > &e, ComputeGraphVar< ViewT > &... vars) |
|
void | when (cudaEvent_t e, int flag=cudaEventWaitDefault) |
|
void | wait (cudaEvent_t e, int flag=cudaEventWaitDefault) |
|
void | wait (const ComputeGraphVar< cudaEvent_t > &e, const std::vector< ComputeGraphVarBase * > &vars) |
|
template<typename... ViewT> |
void | wait (const ComputeGraphVar< cudaEvent_t > &e, ComputeGraphVar< ViewT > &... vars) |
|
void | wait () |
|
void | callback (const std::function< void(::cudaStream_t, ::cudaError)> &callback) |
|
template<typename... ViewT> |
MUDA_INLINE void | record (ComputeGraphVar< cudaEvent_t > &e, ComputeGraphVar< ViewT > &... vars) |
|
template<typename... ViewT> |
MUDA_INLINE void | wait (const ComputeGraphVar< cudaEvent_t > &e, ComputeGraphVar< ViewT > &... vars) |
|