Issue No.01 - Jan.-June (2014 vol.13)
Nadav Rotem , Computer Science Department, Haifa University
Yosi Ben Asher , Computer Science Department, Haifa University
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/L-CA.2012.28
Graphics Processing Units accelerate data-parallel graphic calculations using wide SIMD vector units. Compiling programs to use the GPU's SIMD architectures require converting multiple control flow paths into a single stream of instructions. IF-conversion is a compiler transformation, which converts control dependencies into data dependencies, and it is used by vectorizing compilers to eliminate control flow and enable efficient code generation. In this work we enhance the IF-conversion transformation by using a block unification method to improve the currently used block flattening method. Our experimental results demonstrate that our IF-conversion method is effective in reducing the number of predicated instructions and in boosting kernel execution speed.
Vectors, Merging, Computer architecture, Registers, Kernel, Graphics processing unit,Code generation, Software/Software Engineering, Programming Languages, Processors, Compilers, Software/Software Engineering, Programming Languages, Processors
Nadav Rotem, Yosi Ben Asher, "Block Unification IF-conversion for High Performance Architectures", IEEE Computer Architecture Letters, vol.13, no. 1, pp. 17-20, Jan.-June 2014, doi:10.1109/L-CA.2012.28