gelu_tile

template<bool fast_and_approx = true, bool is_fp32_dest_acc_en = DST_ACCUM_MODE>
void ckernel::gelu_tile_init()

Please refer to documentation for any_init.

template<bool fast_and_approx = true, bool is_fp32_dest_acc_en = DST_ACCUM_MODE>
void ckernel::gelu_tile(uint32_t idst)

Performs element-wise computation of gelu on each element of a tile in DST register at index tile_index. The DST register buffer must be in acquired state via acquire_dst call. This call is blocking and is only available on the compute engine.

Return value: None

Argument Description

tile_index The index of the tile in DST register buffer to perform the computation on

fast_and_approx Computation to be done faster and approximate