Mojo function

blockwise_scaled_fp8_with_epilogue

blockwise_scaled_fp8_with_epilogue[c_type: DType, a_type: DType, b_type: DType, a_scales_type: DType, b_scales_type: DType, //, *, scales_granularity_mnk: IndexList[3], transpose_b: Bool = False, elementwise_lambda_fn: Optional[def[dtype: DType, width: Int, *, alignment: Int = 1](IndexList[2], SIMD[dtype, width]) capturing -> None] = None](c: NullableTileTensor[c_type, c.LayoutType, c.origin, linear_idx_type=c.linear_idx_type, element_size=c.element_size], a: TileTensor[a_type, a.LayoutType, a.origin, linear_idx_type=a.linear_idx_type, element_size=a.element_size], b: TileTensor[b_type, b.LayoutType, b.origin, linear_idx_type=b.linear_idx_type, element_size=b.element_size], a_scales: TileTensor[a_scales_type, a_scales.LayoutType, a_scales.origin, linear_idx_type=a_scales.linear_idx_type, element_size=a_scales.element_size], b_scales: TileTensor[b_scales_type, b_scales.LayoutType, b_scales.origin, linear_idx_type=b_scales.linear_idx_type, element_size=b_scales.element_size], ctx: DeviceContext)

Our sm100 blockwise scaled fp8 matmul kernel still does not support fusion of elementwise operations. This is a temporary implementation that uses our sm100 blockwise scaled fp8 matmul kernel and dispatch a separate epilogue kernel to apply the elementwise operations. For non B200 GPUs, we use the naive blockwise scaled fp8 matmul which support normal epilogue natively.