Skip to main content

Mojo function

mla_prefill_plan_kernel

mla_prefill_plan_kernel[BufferRowOffsetsLayoutType: TensorLayout, CacheOffsetsLayoutType: TensorLayout, BufferLengthsLayoutType: TensorLayout, InputRowOffsetsLayoutType: TensorLayout, cache_t: KVCacheT](buffer_row_offsets: TileTensor[DType.uint32, BufferRowOffsetsLayoutType, MutExternalOrigin], cache_offsets: TileTensor[DType.uint32, CacheOffsetsLayoutType, MutExternalOrigin], buffer_lengths: TileTensor[DType.int32, BufferLengthsLayoutType, MutExternalOrigin], input_row_offsets: TileTensor[DType.uint32, InputRowOffsetsLayoutType, ImmutExternalOrigin], k_cache: cache_t, buffer_token_size: UInt32)

Was this page helpful?