/home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-composable-kernel/checkouts/develop/include/ck/tensor_operation/gpu/grid/gridwise_softmax.hpp File Reference

/home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-composable-kernel/checkouts/develop/include/ck/tensor_operation/gpu/grid/gridwise_softmax.hpp File Reference#

Composable Kernel: /home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-composable-kernel/checkouts/develop/include/ck/tensor_operation/gpu/grid/gridwise_softmax.hpp File Reference
gridwise_softmax.hpp File Reference

Go to the source code of this file.

Classes

struct  ck::GridwiseSoftmax_mk_to_mk< InDataType, OutDataType, AccDataType, GridDesc_M_K, BlockSize, MThreadClusterSize, KThreadClusterSize, MThreadSliceSize, KThreadSliceSize, InSrcVectorDim, InSrcVectorSize, OutDstVectorSize, SweepOnce >
 

Namespaces

 ck
 

Functions

template<typename GridwiseReduction , typename InDataType , typename OutDataType , typename AccDataType , typename GridDesc_M_K >
__global__ void ck::kernel_softmax (const GridDesc_M_K in_grid_desc_m_k, const GridDesc_M_K out_grid_desc_m_k, index_t block_group_size, index_t num_k_block_tile_iteration, AccDataType alpha, const InDataType *const __restrict__ p_in_value_global, AccDataType beta, OutDataType *const __restrict__ p_out_value_global)