A model for fusion and code motion in an automatic parallelizing compiler
Uday Kumar Reddy Bondhugula, Oktay Günlük, Sanjeeb Dash, Lakshminarayanan Renganarayanan · 2010
Loop fusion has been studied extensively, but in a manner isolated from other transformations. This was mainly due to the lack of a powerful intermediate representation for application of compositions of high-level transformations. Fusion presents strong interactions with parallelism and locality. Currently, there exist no models to determine good fusion structures integrated with all components of an auto-parallelizing compiler. This is also one of the reasons why all the benefits of optimization and automatic parallelization of long sequences of loop nests spanning hundreds of lines of code have never been explored.