Domain-specific library generation for parallel software and hardware platforms
Franz Franchetti, Yevgen Voronenko, Peter A. Milder, Srinivas Chellappa, Marek R. Telgarsky, Hao Yang Shen, Paolo D’Alberto, Frédéric de Mesmay, James C. Hoe, José M. F. Moura, Markus Püschel · 2008
We overview a library generation framework called Spiral. For the domain of linear transforms, Spiral automatically generates implementations for parallel platforms including SIMD vector extensions, multicore processors, field-programmable gate arrays (FPGAs) and FPGA accelerated processors. The performance of the generated code is competitive with the best available hand-written libraries.