POLSCA: Polyhedral High-Level Synthesis with Compiler Transformations

被引：3

作者：

Zhao, Ruizhe ^{[1
]}

Cheng, Jianyi ^{[2
]}

Luk, Wayne ^{[1
]}

Constantinides, George A. ^{[2
]}

机构：

[1] Imperial Coll London, Dept Comp, London, England

[2] Imperial Coll London, Dept Elect & Elect Engn, London, England

来源：

2022 32ND INTERNATIONAL CONFERENCE ON FIELD-PROGRAMMABLE LOGIC AND APPLICATIONS, FPL | 2022年

基金：

英国工程与自然科学研究理事会;

关键词：

polyhedral model; high-level synthesis; compiler;

D O I：

10.1109/FPL57034.2022.00044

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Polyhedral optimization can parallelize nested affine loops for high-level synthesis (HLS), but polyhedral tools are HLS-agnostic and can worsen performance. Moreover, HLS tools require user directives which can produce unreadable polyhedral-transformed code. To address these two challenges, we present POLSCA, a compiler framework that improves polyhedral HLS workflow by automatic code transformation. POLSCA decomposes a design before polyhedral optimization to balance code complexity and parallelism, while revising memory interfaces of polyhedral-transformed code to make partitioning explicit for HLS tools; it enables designs to benefit more easily from polyhedral optimization. Experiments on Polybench/C show that POLSCA designs are 1.5 times faster on average compared with baseline designs generated directly from applying HLS on C code.

引用

页码：235 / 242

页数：8