Performance Portability Across Heterogeneous SoCs Using a Generalized Library-Based Approach.
Shuangde FangZidong DuYuntan FangYuanjie HuangYang ChenLieven EeckhoutOlivier TemamHuawei LiYunji ChenChengyong WuPublished in: ACM Trans. Archit. Code Optim. (2014)