We explore the utilization of the Apache TVM open source framework to automatically generate a family of algorithms that follow the approach taken by popular linear algebra libraries, such as GotoBLAS2, BLIS, and OpenBLAS, to obtain high-performance blocked formulations of the general matrix multiplication (gemm). In addition, we fully automatize the generation process by also leveraging the Apache TVM framework to derive a complete variety of the processor-specific micro-kernels for gemm. This is in contrast with the convention in high-performance libraries, which hand-encode a single micro-kernel per architecture using Assembly code. In global, the combination of our TVM-generated blocked algorithms and micro-kernels for gemm (1) improves portability, maintainability, and, globally, streamlines the software life cycle; (2) provides high flexibility to easily tailor and optimize the solution to different data types, processor architectures, and matrix operand shapes, yielding performance on a par (or even superior for specific matrix shapes) with that of hand-tuned libraries; and (3) features a small memory footprint.

我们探索利用 Apache TVM 开源框架自动生成一系列算法，这些算法遵循流行的线性代数库（例如 GotoBLAS2、BLIS 和 OpenBLAS）所采用的方法，以获得通用矩阵乘法的高性能分块公式（宝石）。此外，我们还利用 Apache TVM 框架来完全自动化生成过程，为gemm派生出各种特定于处理器的微内核。这与高性能库中的惯例形成鲜明对比，高性能库使用汇编代码对每个架构手动编码单个微内核。在全球范围内，我们的 TVM 生成的阻塞算法和gemm的微内核的结合(1) 提高了可移植性、可维护性，并且在全球范围内简化了软件生命周期；(2) 提供高度的灵活性，可以轻松地针对不同的数据类型、处理器架构和矩阵操作数形状定制和优化解决方案，从而产生与手动调整库相当的性能（对于特定矩阵形状甚至更优）；(3) 内存占用小。