cub::DeviceFind#

struct DeviceFind#

Public Static Functions

template<typename InputIteratorT, typename OutputIteratorT, typename ScanOpT, typename NumItemsT>
static inline cudaError_t FindIf(
void *d_temp_storage,
size_t &temp_storage_bytes,
InputIteratorT d_in,
OutputIteratorT d_out,
ScanOpT scan_op,
NumItemsT num_items,
cudaStream_t stream = 0
)#

Finds the first element in the input sequence that satisfies the given predicate.

  • The search terminates at the first element where the predicate evaluates to true.

  • The index of the found element is written to d_out.

  • If no element satisfies the predicate, num_items is written to d_out.

  • The range [d_out, d_out + 1) shall not overlap [d_in, d_in + num_items) in any way.

  • When d_temp_storage is nullptr, no work is done and the required allocation size is returned in temp_storage_bytes. See Determining Temporary Storage Requirements for usage guidance.

Snippet#

The code snippet below illustrates the finding of the first element that satisfies the predicate.

Template Parameters:

  • InputIteratorT – [inferred] Random-access input iterator type for reading input items (may be a simple pointer type)

  • OutputIteratorT – [inferred] Random-access output iterator type for writing the result index (may be a simple pointer type)

  • ScanOpT – [inferred] Unary predicate functor type having member bool operator()(const T &a)

  • NumItemsT – [inferred] An integral type representing the number of input elements

Parameters:

  • d_temp_storage – [in] Device-accessible allocation of temporary storage. When nullptr, the required allocation size is written to temp_storage_bytes and no work is done.

  • temp_storage_bytes – [inout] Reference to size in bytes of d_temp_storage allocation

  • d_in – [in] Random-access iterator to the input sequence of data items

  • d_out – [out] Random-access iterator to the output location for the index of the found element

  • scan_op – [in] Unary predicate functor for determining whether an element satisfies the search condition

  • num_items – [in] Total number of input items (i.e., the length of d_in)

  • stream – [in]

    [optional] CUDA stream to launch kernels within. Default is stream0.

On this page