Mojo function
grouped_smem_size
grouped_smem_size[a_type: DType, b_type: DType, c_type: DType, sfa_dtype: DType, sfb_dtype: DType, transpose_b: Bool, config: BlockScaledMatmulConfig[a_type, b_type, c_type, sfa_dtype, sfb_dtype, transpose_b]]() -> Int
Calculate shared memory size for grouped block-scaled kernel.
Returns:
Int: SMEM size in bytes, including tensormap descriptor storage.
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!